Common source bias in ads

Status
Not open for further replies.
S

Sarvenaz

Junior Member level 1
Joined
Dec 2, 2015
Messages
16
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
110
Hi
How can i bias a nmos 2v in 0.18um in ads?
 

Hi
When i name VD=Vo and simulate it
The Vo(dB) becomes -1
 

Sarvenaz said:
When i name VD=Vo and simulate it
The Vo(dB) becomes -1
Click to expand...
Show me ADS netlist.

I think you don't set small signal amplitude for "VtSine".

If you can not set small signal amplitude for "VtSine", use "V1_tone" or "V_AC" instead of "VtSine".

Note "VtSine", "V1_tone" and "V_AC" are all same component "V_Source" in netlist.
Just they are different regarding parameters.
 
Last edited:
Sarvenaz said:
Hi
When i name VD=Vo and simulate it
The Vo(dB) becomes -1
Click to expand...

since you are not doing DC simulation, that is AC result. ADD simulation-DC to see DC result.

- - - Updated - - -

The source you put is for transient simulation and you are doing AC simulation. how can you sweep frequency and your source frequency is 10 MHz ? Have you done AC simulation before ?

Pick an AC voltage source you dont have AC value there.

Add a source from this library Sources-Freq Domain
 
Last edited:
Ata_sa16 said:
since you are not doing DC simulation, that is AC result.
ADD simulation-DC to see DC result.
Click to expand...
DC operation points are calculated in all analyses of ADSsim.
You can see DC operation point's information from AC analysis.
So there is no need to place DC-Analysis controler.
 
Yeah that is true. But if you want to Annotate DC solution you need DC simulation. I though he may need to see bias point on circuit.

I have found his problem, I explained it.
 

pancho_hideboo said:
No.
ADSsim does not require DC Analysis.

On the other hand, Cadence Spectre requires DC Analysis.
Click to expand...

Do you know how to see bias points on circuit ? I am talking about Annotate DC solution and it needs DC simulation. ADSsim does not require DC analysis but for "Annotate DC solution" you MUST do DC analysis.

I did not say your simulation does not work or does not do DC simulation, I offered him to do a separate DC analysis in order to observe his problem better; however, problem was other thing.

I realized he is talking about bias and he is doing AC simulation, I offered him to use only DC simulation to focus on biasing otherwise I never said for AC simulation you need DC analysis as well.
 
Last edited:

Ata_sa16 said:
Do you know how to see bias points on circuit ?
I am talking about Annotate DC solution and it needs DC simulation.
Click to expand...
It seems you are very beginner regarding simulator.
See https://www.designers-guide.org/Forum/YaBB.pl?num=1212376329/1#1
I don't place DC-Analysis there.

Ata_sa16 said:
ADSsim does not require DC analysis but for "Annotate DC solution" you MUST do DC analysis.
Click to expand...
Wrong.
Post after you can surely study simulator.

Simply you are offering unnecessary informations.

If you are using ADSsim in Cadence Virtuoso, DC Analysis is required for annotation of operation points.
This is because PSF data have to be created by DC-Analysis in Cadence Virtuoso.
 
Last edited:
pancho_hideboo said:
It seems you are very beginner regarding simulator.
Click to expand...


This is a disrespectful comment. This is not a correct commenting manner and attitude. I am not a beginner I designed many ICs with ADS.

Probably I forgot about that since I always work with Harmonic balance simulations and sometimes in those simulations if you dont put DC analysis you wont see bias point. Thank you for clarification.

- - - Updated - - -


Remember your comment, you had not realized he is putting a wrong source and I never called you beginner, everyone makes mistake. you are a professinal ADS user you should now that for AC simulation you need freq domain source not vtsine.
 
Ata_sa16 said:
I am not a beginner I designed many ICs with ADS.
Click to expand...
I don't think so.
You are no more than student level.
https://www.edaboard.com/showthread.php?t=356560

Ata_sa16 said:
Probably I forgot about that
since I always work with Harmonic balance simulations
and sometimes in those simulations if you dont put DC analysis you wont see bias point.
Click to expand...
Wrong.
DC operation point's annotation is available even in HB-Analysis.

DC operation point's annotation is available in Any Analysis of ADSsim, since any analysis starts with determining DC operation point.

On the other hand, HSPICE require ".OP" analysis and Cadence Spectre require "dc" analysis.

Ata_sa16 said:
for AC simulation you need freq domain source not vtsine.
Click to expand...
Surely see ADS netlist.
There is no "VtSine" as Netlist expression.

Any voltage sources are described by "V_Source".

Any voltage source in component palette is no more than symbol to generate parameters for "V_Source".

This is same as "vsource" in Cadence Spectre.
 
Last edited:
It is not important what you think. I am not a student but I am teacher.

The post you mentioned is about VCO design (July 2016) since that time I design 2 LNA UWB and 1 PA ICs, I dont have experience with VCO design.

You never gain anything by humiliating people.

- - - Updated - - -

Sarvenaz said:
Hi
How can i bias a nmos 2v in 0.18um in ads?View attachment 140155
Click to expand...


How can you do AC simulation with putting vtsine in this schematic with constant frequency ?

He has used vtsine voltage source which is used for transient simulation. I tell him change it you call me student !
 
Last edited:
Sarvenaz said:
How can i bias a nmos 2v in 0.18um ... ?
Click to expand...

Finding the right Vgs bias may become a tedious try-and-error task; I think it's much better to define this bias by means of a current mirror.

You define the current Ids(X1) for opt. modulation range, considering that the transistor should always operate in saturation region. To be on the safe side, take Vdsat ≥ Vth , say Vdsat=0.8V. Then

V(R1) = (V_DC - Vdsat)/2 = 0.5V
Vds(X1) = (V_DC - Vdsat)/2 + Vdsat = (V_DC + Vdsat)/2 = 1.3V

I(R1) = Ids(X1) = V(R1)/R1 = 0.5mA
Click to expand...

Use this current for a 1:1 current mirror with NFET X2 (or a scaled-down current with an equally scaled-down X2) and you get automatically the right Vgs bias voltage for X1. You have to separate this bias voltage from the ac stimulation voltage by a capacitor, otherwise the ac source would short-circuit the DC bias.



The midband voltage gain Av ≈ -4 . The decay towards lower frequencies is due to C1 * Rout(X2) .
 





This will help you.

I have not designed this, I just put some random values in order to solve your simulation problem. I dont want people come and say why the gain is 2 and stuff like this
 

Sarvenaz said:
Hi
When i name VD=Vo and simulate it
The Vo(dB) becomes -1
Click to expand...

For able to do AC simulation, your source has to have a field "AC Voltage" otherwise your simulation will not work.You're using Vtsine source that is a special source for different simulation purpose.
By the way, why AC simulation you intend to do ?? Why don't you simply do a first transient simulation then HB simulation ?? So you obtain all necessary information for your circuit.
 

Any voltage source in pallet of ADS is "V_Source" as netlist expression.
Code: 
% adssim -h V_Source

V_Source [:Name] p n <parameter=value> ... ; (device)
Parameters:
    Vdc (V)          smorr DC voltage.
    Vac (V)          smorc AC voltage.
    V[,,] (V)        smorc Large signal voltage at a given harmonic.
    V_All            smorc Large signal voltage at all harmonics.
    V_USB[,,] (V)    smo-c Small signal voltage at a given upper sideband.
    V_LSB[,,] (V)    smo-c Small signal voltage at a given lower sideband.
    FundIndex        sm--i Single fundamental index for this source.
    V_Noise (V)      smorr Noise Voltage.
    PhaseNoise (dB)  smo-r Phase noise as a function of offset frequency.
    V_Tran (V)       smorr Transient voltage.
    HarmList (V)     smo-c Array of large signal harmonics.
    Freq (Hz)        smo-r Actual n'th frequency value.
    SaveCurrent      s--rb Send source current to rawfile.
    Tnom (deg C)     smorr Nominal temperature in degrees Celsisus.
    TC1 (1/deg C)    smorr Linear temperature coefficient; per degree Celsius.
    TC2 (1/(deg C)^2) smorr Temperature coefficient; per degree Celsius squared.
    Type             s--rs Source Type (e.g. V_DC for DC source) genereated by PDE..

Instance DC operating point parameters:
    Is (A)           ---rr Current.
    Power (W)        ---rr DC power dissipated.
    Vs (V)           ---rr Voltage.

We have to set small signal amplitide, "Vac" as non-zero value for AC analysis.
"Freq" parameter has no meaning for conventional AC analysis.
So you can set any value as "Freq" for conventional AC analysis.
Or you don't have to set "Freq" for conventional AC analysis.

Mandatory parameter for conventional AC analysis is "Vac".
"Freq" is not required at all.

Ata_sa16 said:
How can you do AC simulation with putting vtsine in this schematic with constant frequency ?
Click to expand...
You can not understand my following append.
Some sources in time domain pallet have "Vac" parameter.
For example, followings have "Vac" parameter.
"V_DC (DC Voltage Source, Time Domain)"
"VtUserDef (Voltage Source, User-Defined)"

However, to my regret, "VtSine" does not have "Vac" parameter.
So we can not set "Vac" from GUI.
Of course, you can add "Vac" to line for "VtSine" in netlist by yourself.
There is no "VtSine" in netlist. Netlist expression is "V_Source".

Sarvenaz use "V_DC" and "VtSine" as series connection.
"V_DC" has "Vac" parameter, so you can set small signal amplitude for "V_DC".
So there is no need to replace component symbol.
 
Last edited:

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…