Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Why does the ac curve of this OPA has no -3dB breakoff point

Status
Not open for further replies.

huojinsi

Full Member level 5
Joined
Jan 21, 2005
Messages
256
Helped
25
Reputation
50
Reaction score
6
Trophy points
1,298
Activity points
2,216
Hi every one:
I am learning a schematic of OPA which is quoted from a textbook. When i simulated the open loop gain frequency curve, i found the gain frequency curve of open loop didn't degrade with frequency varying, In other words, the gain curve doesn't have -3db breakoff point when the frequency is increasing. I am very confused, because the models file contain capacitor paramter and the circuit contain compensation capacitor. Why to yield this strange phenomenon? Pls guide me! If u have a free time, pls help me simulate it! The schematic of OPA is in the attachment.Thks in advanced!
my models file is
"*FOR SPECTRE
.model MOSN NMOS
+Level= 11
*
* GENERAL PARAMETERS
*
+lmin=6.0e-7 lmax=3.0e-5
+wmin=6.0e-7 wmax=6.0e-5
+Tnom=25.0 version=3.1
+Tox=1.25E-08 Xj=1.0000000E-07
+Nch=1.3196000E+17 lln=0.7550657
+lwn=1.0000000 wln=1.7385179
+wwn=2.0000000 lint=1.6723000E-07
+ll=-1.6375242E-12 lw=0.00
+lwl=-4.0506720E-19 wint=1.2458000E-07
+wl=-1.8283731E-19 ww=-2.3537305E-20
+wwl=0.00 Mobmod=1
+binunit=2 xl=0.00
+xw=0.00 Dwg=-1.5901913E-08
+Dwb=1.4839435E-08
*
* DIODE PARAMETERS
*
+ldif=0.00 hdif=0.00
+rsh=0 rd=0
+rs=0 rsc=0
+rdc=0
*
* THRESHOLD VOLTAGE PARAMETERS
*
+Vth0=0.6929701 K1=0.8389516
+K2=-5.7840230E-02 K3=2.5218985
+Dvt0=4.7742720 Dvt1=0.8496043
+Dvt2=-0.1037025 Dvt0w=0.00
+Dvt1w=0.00 Dvt2w=0.1900000
+Nlx=0.00 W0=2.3155847E-07
+K3b=-2.6079214 Ngate=1.0000000E+30
*
* MOBILITY PARAMETERS
*
+Vsat=8.8189170E+04 Ua=-8.8038300E-10
+Ub=2.8512771E-18 Uc=3.6204920E-11
+Rdsw=1.3140129E+03 Prwb=1.8216049E-02
+Prwg=-3.5016470E-02 Wr=0.9094040
+U0=4.2317160E-02 A0=0.2826061
+Keta=-1.4081742E-02 A1=0.00
+A2=0.9900000 Ags=2.0000000E-02
+B0=4.5134230E-08 B1=-3.6182750E-07
*
* SUBTHRESHOLD CURRENT PARAMETERS
*
+Voff=-9.6546230E-02 NFactor=1.6005070
+Cit=-6.5337530E-04 Cdsc=0.00
+Cdscb=0.00 Cdscd=4.6825380E-05
+Eta0=0.1064645 Etab=-5.5450850E-02
+Dsub=0.6590825
*
* ROUT PARAMETERS
*
+Pclm=2.2878227 Pdiblc1=0.1000000
+Pdiblc2=8.7633540E-04 Pdiblcb=-0.1000000
+Drout=0.4651489 Pscbe1=4.8989800E+08
+Pscbe2=1.0000000E-20 Pvag=1.0000000
+Delta=1.0000000E-02 Alpha0=0.00
+Beta0=30.0000000
*
* TEMPERATURE EFFECTS PARAMETERS
*
+kt1=-0.3654486 kt2=-2.5749199E-02
+At=1.5000000E+04 Ute=-1.3589500
+Ua1=2.7960341E-09 Ub1=-3.4426440E-18
+Uc1=-1.8533992E-11 Kt1l=0.00
+Prt=0.00
*
* CAPACITANCE PARAMETERS
*
+Cj=0.000343 Mj=0.593
+Pb=2.55 Cjsw=2.91E-09
+Mjsw=0.362 Pbsw=0.303
+Cjswg=2E-14 Mjswg=0.33
+Pbswg=0.7 Cta=0.00105
+Ctp=-2.90E-04 Pta=-3.59E-04
+Ptp=5.06E-4 JS=1.00E-04
+JSW=0.00 N=1.0
+Xti=3.0 Cgdo=9.0035E-11
+Cgso=3.7503E-10 Cgbo=1E-12
+Capmod=3 NQSMOD=0
+Elm=5 Xpart=0
+cgsl=5.0000000E-11 cgdl=4.1552E-10
+ckappa=0.6000000 cf=0.00
+clc=8.56000E-09 cle=0.8100000
+Dlc=-5.79E-07 Dwc=1.9992E-09
+Vfbcv=-0.838282
*
*
***************************************************************
* PMOS DEVICES MODEL *
***************************************************************
* FOR SPECTRE
.model MOSP PMOS
+Level= 11
*
* GENERAL PARAMETERS
*
+lmin=6.0e-7 lmax=3.0e-5
+wmin=6.0e-7 wmax=6.0e-5
+Tnom=25.0 version=3.1
+Tox=1.25E-08 Xj=1.0000000E-07
+Nch=3.8952000E+16 lln=0.3728379
+lwn=1.0000000 wln=1.9910562
+wwn=1.3712339 lint=-7.8715000E-08
+ll=2.0611995E-10 lw=0.00
+lwl=-1.0559602E-16 wint=2.2922001E-07
+wl=-9.4221030E-22 ww=-3.9271670E-16
+wwl=0.00 Mobmod=1
+binunit=2 xl=0.00
+xw=0.00 Dwg=-1.7118449E-08
+Dwb=9.4421320E-09
*
* DIODE PARAMETERS
*
+ldif=0.00 hdif=0.00
+rsh=0 rd=0
+rs=0 rsc=0
+rdc=0
*
* THRESHOLD VOLTAGE PARAMETERS
*
+Vth0=-0.9534206 K1=0.4306269
+K2=2.4288371E-02 K3=2.8558486
+Dvt0=9.2671260 Dvt1=1.2710701
+Dvt2=-0.1255428 Dvt0w=0.00
+Dvt1w=0.00 Dvt2w=0.1900000
+Nlx=0.00 W0=-1.8608442E-07
+K3b=-0.6373694 Ngate=1.0000000E+30
*
* MOBILITY PARAMETERS
*
+Vsat=1.5532217E+05 Ua=2.3390669E-09
+Ub=2.9231469E-19 Uc=-4.1082440E-11
+Rdsw=1.2391572E+03 Prwb=-2.5805227E-02
+Prwg=-8.4016780E-03 Wr=0.8132943
+U0=2.1171642E-02 A0=0.4847234
+Keta=4.2094130E-03 A1=0.00
+A2=0.4000000 Ags=2.0000000E-02
+B0=1.5313557E-07 B1=-3.1252178E-08
*
* SUBTHRESHOLD CURRENT PARAMETERS
*
+Voff=-0.1121503 NFactor=2.0000000
+Cit=-4.0293170E-04 Cdsc=4.1957440E-04
+Cdscb=0.00 Cdscd=4.3705550E-05
+Eta0=0.2910967 Etab=-2.8794665E-02
+Dsub=0.6654378
*
* ROUT PARAMETERS
*
+Pclm=5.0000000 Pdiblc1=1.0000000E-05
+Pdiblc2=1.5418571E-03 Pdiblcb=0.1000000
+Drout=3.0000000 Pscbe1=4.8989800E+08
+Pscbe2=1.0000000E-20 Pvag=0.8505571
+Delta=1.0000000E-02 Alpha0=0.00
+Beta0=30.0000000
*
* TEMPERATURE EFFECTS PARAMETERS
*
+kt1=-0.4858726 kt2=-2.0427300E-02
+At=1.0000000E+04 Ute=-1.0726622
+Ua1=3.7500000E-09 Ub1=-6.6092640E-18
+Uc1=-7.5526270E-11 Kt1l=0.00
+Prt=0.00
*
* CAPACITANCE PARAMETERS
*
+Cj=0.000704 Mj=0.426
+Pb=0.855 Cjsw=2.81E-10
+Mjsw=0.804 Pbsw=1.53
+Cjswg=2E-14 Mjswg=0.33
+Pbswg=0.7 Cta=0.000947
+Ctp=0.00284 Pta=2.17E-03
+Ptp=5.26E-03 JS=1.00E-04
+JSW=0.00 N=1.0
+Xti=3.0 Cgdo=1.2004E-10
+Cgso=8.2544E-11 Cgbo=1E-12
+Capmod=3 NQSMOD=0
+Elm=5 Xpart=0
+cgsl=8.0000000E-13 cgdl=3.67460000E-10
+ckappa=0.6000000 cf=0.00
+clc=1.0000000E-09 cle=0.600000
+Dlc=-2.98E-07 Dwc=0
+Vfbcv=0.07944
*
"
 

i have a question where is the bias of M1.
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

A better way to test this circuit is to use stability analysis.

Just connect the Opamp in umity gain buffer configuration. In the feedback path insert an "iprobe" in series. you will find it in analogLib.

After running stability analysis, you will be able to plot phase and gain curves. Also, Cadence automatically calculates the Phase and gain margins..

This saves you the fuss of finding a good way to break the loop while estabilishing a good dc-bias.

regards
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

safwatonline said:
i have a question where is the bias of M1.
it has connected as unit buffer, so M1 need not bias.
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

Check if this
W=12.5
L=1
should be:
W=12.5u
L=1u

If you want to see something from amplitude and phase characteristics you should set x axes view as log (Axis meny).
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

mists said:
it has connected as unit buffer, so M1 need not bias.
yes, i see the coil but what makes the dc of M1 equal to M2
i think that as elbadry said stability analysis is better

pixel said:
Check if this
W=12.5
L=1
should be:
W=12.5u
L=1u

If you want to see something from amplitude and phase characteristics you should set x axes view as log (Axis meny).

yes , this seems a problem
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

safwatonline said:
yes, i see the coil but what makes the dc of M1 equal to M2
i

It makes negative feedback. Coil is short circuit at dc.
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

ok ,what i mean is that in open loop simulation , he cuts the loop then what is the dc of M1 . i think stability analysis will solve this.

btw does the DC of M1 = DC of M2 or not , if not what makes the diff. pair symmetric?
thnx
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

safwatonline said:
ok ,what i mean is that in open loop simulation , he cuts the loop then what is the dc of M1 . i think stability analysis will solve this.

btw does the DC of M1 = DC of M2 or not , if not what makes the diff. pair symmetric?
thnx

Yes of course. In open loop simulation both dc voltages has to be equal.
 

    huojinsi

    Points: 2
    Helpful Answer Positive Rating
Please set the x axes of AC responce in log scale
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

I think the value of capacitor should be 1G.
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

vincentren said:
I think the value of capacitor should be 1G.
Yes L=10G and C=1G !?... How much m² would be your chip ?
 

Re: Why does the ac curve of this OPA has no -3dB breakoff p

Dear huojinsi,

I have few comments here:

1. Are you sure your MOS size is correct? should it in micro meter range? please check and confirm with us

2. Can you please change your Capacitor value C1=1GF and instead of putting the Acm=1 at V2, creat another Vdc with dc=0, ac=1 (V2 with dc=5v only)and connect it btw C1 and GND.

3. If you still cannot get it, can you annotate all the DC point and show it again.




Good luck

Regards,
Smart

Added after 1 minutes:


Yes L=10G and C=1G !?... How much m² would be your chip ?[/quote]


THis is so-called Middlebrook setup, it is not going to intergrate on chip, just for verificatio purpose.


Regards,
Smart
 

    huojinsi

    Points: 2
    Helpful Answer Positive Rating
Re: Why does the ac curve of this OPA has no -3dB breakoff p

especially thks pixel and always@smart! very appreciate ur mind!
I make a mistake, and i lose the "u" unit of W and L. Now, i correct them and the right simulation result has came out!
Regarding to L=10G, only to acheive blocking of ac signal, thus this circuit is a equivalent open loop state, so that the open loop gain is simulated.

Thks eveyone again!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top