Explain Calculation of power in analog ICs Vs Digital ICs

[sandeep_sggs]

If we talk about digital IC, the power consumption is classified as:
STATIC POWER
DYNAMIC POWER

But if we refer to any analog IC related papers, only single power is mention for any relevant circuit like 0.25mW, 4mW, 0.6mW (milli Watt), etc..Can anybody explain why is it so?

 

[erikl]

... if we refer to any analog IC related papers, only single power is mention for any relevant circuit like 0.25mW, 4mW, 0.6mW (milli Watt), etc..Can anybody explain why is it so?

If so, always the analog cell's power consumption in its actual application is meant. Could be STATIC POWER (e.g. for a current mirror) or DYNAMIC POWER for converters or RF circuits. In such case the operating frequency has to be stated, too, because these circuits' power consumption depends on it.

 

[sandeep_sggs]

thank you erikl for reply,
i have attached a part of some analog paper. The table shows the power in mW. This is a comparative table showing various CMOS OPAmp compensation topologies. This paper has directly computed power by multiplying vdd(supply voltage) with the current (Bias current). This computation of power is not shown in table! My question is
1) will it not depend on operating frequency?
2) Why it has computed only DC power i.e vdd*Idrawn from supply
I have simulated the circuits. The transient plot of power drawn from vdd has both DC as well as AC component.

3) Why no mention of AC power component?

 

[erikl]

1) Sure it will. As well as on output capacitance.
2) DC power (or quiescent power) is given without input activation (no frequency input, no clock). Idrawn is the quiescent consumption current.
With input/clock activation you have an additional ac component, mainly used for charging/discharging internal and external capacitances.
3) Your table contains operation frequencies, so I suppose the given power should include DC+ac power at the given frequency and for the given load capacitance of C_L=30pF.

 

[sandeep_sggs]

i think frequency that is mentioned in the table is unity gain frequency of the opamp. It(Fun) tells us about the maximum speed with which the op-amp can operate. The power mentioned there are just multiplication of Vdd and Idrawn from supply, as you have pointed out, so called DC power (or quiescent power).

When i try to plot power for DC source (Vdd), i get straight line i.e. a DC power + super-imposed on DC i get AC power component which is sinusoidally(since input is small signal AC signal) varying.

My question is still the same why no mention of AC power component, is it because the DC power is a big quantity as compared to AC power and that small signal analysis is followed(done) here?

 

[erikl]

... why no mention of AC power component, is it because the DC power is a big quantity as compared to AC power and that small signal analysis is followed(done) here?

It ain't necessarily so ;-). Ac power at operation frequency may be small, comparable to or even considerably bigger than the quiescent (DC) power - depending on if it is a small signal, medium or high power amplifier, at least for a well designed amp. Amps with class A output usually show constant power consumption independent of input/clock activation.

In the end you have to ask the author(s) of the paper.