Referencing Wild Card in Search/Replace in Virtuoso Schematic

Here, the {1} is referring to the first wildcard in the search pattern

SFG of A Source

SFG of a source is shown below:

N.B.  if Γs is 0 (which means the source impedance is Zo or 50Ωs), then

dog bone devices

Layout Purpose Synenim

label: ll

pn: pin

osc

RC Oscillator

One form of Schmitt Trigger

Modify Master Value of a Cell

in symbol editor

or in symbol editor press Shift+Q

Net Expressions

Calculate Power Waves and Generalized S-Parameters

In ADS, build a circuit and set parameters as shown below:

and simulate the circuit in 1GHz single point.


After simulation, add equations to get incident and reflected power wave:

Note, ZG and ZL should be added to the output database (in simulation control component)


The generalized S-Parameters can be readily calculated by S_Param simulation.

To verify the theory of power wave, calculate:



The results are totally in agreement with each other. Thus the power wave theory is verified.

Bidirectional Buffer

dafasdfdaf

ESD Circuitry

stub

Razavi Problem 14.5 revisit

To re-approach this problem:

Assume Vi point is driven by a step voltage, the response of Vx to that step can be proved to be as shown in the blue curve. As can be seen, at the onset of the response, it is like a ramp, this ramping of Vx introduces a ramping in the drain current at node Vi, so a step in voltage results in a ramp in current, that is a clear sign of inductor.

Current Direction of Source (spectre)

Current OUT:  negative
Current IN:      positive

                 

Negative Inductor

negative capacitor is effectively a inductor

purple: current
red: voltage:

Solution to Razavi Book Problem 14.5

Including only the gate-source capacitance of M3, explain under what condition the impedance of the composite load (seen at the drain of M3) becomes inductive.

Find KCL equations:

so the output impedance is:

so the equivalent inductance is:

and the output impedance is always inductive.

A transient simulation is conducted to verify the inductive effect of this load: A input step current is initiated and voltage (Vi) is probed. (a 100fF cap is added to make the inductive effect more prominent)

The driving point (voltage) response is shown below (red curve), the inductive effect can be eminently seen (ringing to step response)

The inductive impedance can be intuitively described (in a hand-waving manner) as follows:

When the voltage of the drain of M3 is moving very slow, the gate voltage of M3 follows suit, and only very a quite amount of current will be following into drain of M3. On the other hand, when the drain of M3 is moving very rapidly, the caps at the gate of M3 actually filters out all the high frequency signals, and the voltage at this node basically stays put, and virtually no current will be flowing into the drain of M3, so that means the higher the frequency, the larger the drain impedance, which means it is inductive.


voltage @ /M3/Gate follows current into /M3/Drain,
Current into C leads voltage @ /M3/Gate ( by 90deg )
Current out /M5/Source follows Current into C
voltage @ /M3/Drain follows current out /M5/Source

In summary, voltage @ /M3/Drain leads current into /M3/Drain by 90deg, which results in inductive impedance.

Begin and End Of a line in Visio

Get The Slope Around A Point In A Trace

You can create a Delta Marker to see the slope of a trace

by default, the slope is shown within the parenthesis, as can be verified by seeing the contents of the Delta Marker:

here %S means slope.

Transient Simulation Convergence Issue

In one of my projects, I bumped into a transient convergence issue: the error message shows something like:

************************************************
Transient Analysis `tran': time = (0 s -> 30 ms)
************************************************
Trying `homotopy = gmin' for initial conditions.
Trying `homotopy = source' for initial conditions.
Trying `homotopy = dptran' for initial conditions..
Trying `homotopy = ptran' for initial conditions..
Error found by spectre during IC analysis, during transient analysis `tran'.
    ERROR (SPECTRE-16080): No DC solution found (no convergence).
The values for those nodes that did not converge on the last Newton iteration are given below.  The manner in which the convergence criteria were not satisfied is also given.
            Failed test: | Value | > RelTol*Ref + AbsTol
 Top 10 Residue too large Convergence failure:
    V(I93.M20.gj1) = 336.667 V
        residue too large: | 1.36002 kA | > 6.8001 A + 1 pA
    V(I93.M9.gj1) = 336.667 V
        residue too large: | 1.36002 kA | > 6.8001 A + 1 pA
    V(I93.M5.gj1) = 336.667 V
        residue too large: | 1.36002 kA | > 6.8001 A + 1 pA
    V(I93.M20.d1) = -666.194 V
        residue too large: | -1.35956 kA | > 6.7978 A + 1 pA
    V(I93.M9.d1) = -666.194 V
        residue too large: | -1.35956 kA | > 6.7978 A + 1 pA
    V(I93.M5.d1) = -666.194 V
        residue too large: | -1.35956 kA | > 6.7978 A + 1 pA
    V(I93.M106.gj1) = 359.453 V
        residue too large: | 1.30055 kA | > 6.50273 A + 1 pA
    V(I93.M107.gj1) = 359.453 V
        residue too large: | 1.30055 kA | > 6.50273 A + 1 pA
    V(I93.M1.gj1) = 359.453 V
        residue too large: | 1.30055 kA | > 6.50273 A + 1 pA
    V(I93.M106.d1) = -743.107 V
        residue too large: | -1.29651 kA | > 6.48254 A + 1 pA

Analysis `tran' was terminated prematurely due to an error.
finalTimeOP: writing operating point information to rawfile.
Error found by spectre during DC analysis, during info `finalTimeOP'.
    ERROR (SPECTRE-16041): Analysis was skipped due to inability to compute operating point.

 I finally solved this issue by making all my power lines to be at 0 voltage at the initial time, after the initial zero for a while, the power lines begin ramping up to the intended value.

When said power signals had been changed this way, the initial DC analysis was harnessed.

Threshold-Voltage Loss Compensation (TLC) Loads In Analog IC

dafadaf

Word "Said" In Patent Paragraphs

"Said" can be used in the context of law as meaning "mentioned before", or in Chinese "该"
"Said" used in this way is an adjective word.

e.g.  The said (上述，该) weapon was later found in the defendant's home

Things You Don't Know About Virtuoso: Select All Using Navigator Assistant

To select all components that belong to same master, just open the Navigator Assistant, and choose Show --> Cells:

All cells that belong to the same master are now grouped together, like shown below:

Now, in the Property Editor Assistant, you can change the property of all the cells that belong to the same master once and for all.

When will adsense user start to receive payments?

According to adsense's policy,   they first have to validate user's earnings for accuracy, a process that takes place the first week of every month, and send the payment a couple weeks later, so if your account balance reaches the payment threshold in February,  you will get your payment at the end of March.

Applying for WordAds failed -- Blogs without "custom domains"

In wordpress, to be eligible to apply WordAds, your blog has to have a custom domain (as shown below), or the application cannot be proceeded.

How to change layout of your blogger

Choose Layout

You can change your layout (like add a ad block) this way

ADS Terminator component in transient simulation

It seems that the reactive part of TERM component is ignored by ADS in transient simulation.

I simulated a very simple schematic:

the Term component here is a complex load.


However, the simulation result shows that the voltage and current are in-phase to each other, that means the reactive part of terminator is ignored

as can be seen, the magnitude of load voltage is 2.0V, and the magnitude of load current is 0.04A, so the effective load is 50Ω, not (50+j*100)Ω.

So, in order to get reactive phenomenon in transient simulation using ADS, you should explicitly add one of the two reactive components.

Marker Label in ViVa

%M   Marker name

%X   X-coordinate

%Y   Y-coordinate

%x    Second X-coordinate for delta markers

%y    Second Y-coordinate for delta markers

%W  Dx

%H   Dy

%S    Slope (Dx/Dy)

%N    Name of the trace

%E    Expression

%F    Frequency value

test No.2

test N0.2

voltage wave amplitudes & normalized voltage wave amplitudes

we should first define Vinc and Vref in such a way that their sum is the voltage, and their difference (divided by reference impedance) is the current..

then the definition of Vinc and Vref (in terms of V0 and I0) comes naturally.



for normalized voltage wave amplitudes: