10 views

Original Title: 1

Uploaded by murthy

- Transistor Level Implementation of CMOS Com Bi National Logic Circuits
- syl
- Verilog codes
- Datasheet(1)
- etd
- Vlsi Lab Report Final
- Ultra-Low Power Digital Subthreshold Logic Circuits
- Redefining CMOS Logic Style for Subthreshold
- Osciladores Ingles 1
- mosfet
- Syllabus Amie Syllabus Section b (EC)
- Low-Power Level Shifter For
- TC4469
- VLSI Ecad LAboratory
- all related questions- with 1 b
- 372fbc17145f4f0c7c8947c7f2522d4d.pdf
- P Channel MOSFET:
- Ut28f256qle Prom
- Si_nc
- Juge_NanoTera_2011.pdf

You are on page 1of 4

Homework 6 -Solution

Spring 2008

Problem 1

A NOR CMOS gate with the device/parasitic parameters below must drive (output to) the inputs of 3

NAND gates (one input on each gate) with the same MOSEFT gate dimensions as the NOR gate.

VDD = 2.0V, Cox = 2fF/m2, CDBp = 1fF, CDBn = 0. 5fF

(W/L)p= 4/0.5, kp =100A/V2, |Vtp| = 0.5V,

(W/L)n= 2/0.5, kn =250A/V2, Vtn = 0.6V,

a)

b)

c)

d)

What is the total parasitic capacitance at the output of the NOR gate?

What is the total load capacitance contributed by the 3 NAND gates?

What is the total capacitance at the output of the NOR gate?

What is the effective resistance between the output and VDD when the NOR gate pulls the output

high? Use the general formula for channel resistance.

e) Combining results from (c) and (d), what is the output RC time constant?

solution

a) A NOR gate has 1 pMOS and 2 NMOS drains at the output node.

Thus, Cpara = CDp + 2CDn

CDp = CGDp + CDBp AGp Cox + CDBp = (4*0.5)(2) + 1 = 3 fF

CDn = CGDn + CDBn AGn Cox + CDBn = (2*0.5)(2) + 0.5 = 1.5 fF

Cpara = 3 + 2(1.5) = 6 fF

b) Each input of a NAND gate goes to 1 pMOS and 1 nMOS transistor. The capacitive load

generated by gate (input) node of these transistors is just CGp + CGn.

Thus, CL = 3 (CGp + CGn) = 3 Cox (AGp + AGn) = 3 (2) (4*0.5 + 2*0.5)

CL = 18fF

c) Cout = Cpara + CL = 6 + 18 = 24fF

d) When the NOR gate pulls its output high, both series pMOS must be on.

Thus, Reff = 2 Rp,

Rp = [p(VDD-|Vtp|)]-1, p = kp (W/L) = 100 (4/0.5) = 800uA/V2

Rp = [800u(2.0-|0.5|)]-1 = 833

e) = Rp Cout = 833 (24f) = 20,000x10-15 = 20x10-12 = 20psec

Problem 2

For the problems below assume the following process parameters and constants: kn = 100 A/V2, kp =

40A/V2, Vtn=|Vtp|=0.5V, VSB = 0V and VDD = 2.5V. Ignore channel length modulation and other

secondary effects.

a) Use a computer (e.g., Excel, Matlab) to plot the I-V curve (ID vs. VD) for an nMOS transistor with

W=1.5m and L=0.6m, VS = 0V at VG = 1V. Plot ID vs. VD from 0V to 2.5V with at least 10

steps. Be sure to properly account for transitions between the triode and saturation (active) regions.

b) Repeat (a) at VG = 2V and VG = 2.5V. Include all three curves on the same plot.

p. 1

c) Repeat (a) and (b) for a pMOS device with W=3m, L=0.6m and VS = 2.5V and VG values of 2V,

1V, and 0V. Notice that to plot ID vs. VD, you will need to be careful how you represent the term

VSD. Done correctly, your plot should be similar to the one shown in HW4 solutions.

Solution

a & b)

Vds

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.5

Vg=1

0.0

20.0

30.0

31.3

31.3

31.3

31.3

31.3

31.3

31.3

31.3

31.3

31.3

31.3

Vg=2

0.0

70.0

130.0

180.0

220.0

250.0

270.0

280.0

281.3

281.3

281.3

281.3

281.3

281.3

c)

Vg=2.5

0.0

95.0

180.0

255.0

320.0

375.0

420.0

455.0

480.0

495.0

500.0

500.0

500.0

500.0

Vsd

2.5

2.3

2.1

1.9

1.7

1.5

1.3

1

0.8

0.6

0.4

0.2

0

0

1

Vg=1

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.0

8.0

0.0

0.0

2

Vg=2

112.5

112.5

112.5

112.5

112.5

112.5

112.5

100.0

88.0

72.0

52.0

28.0

0.0

0.0

2.5

Vg=2.5

200.0

200.0

200.0

199.5

195.5

187.5

175.5

150.0

128.0

102.0

72.0

38.0

0.0

0.0

250.0

600.0

Vg=1

Vg=1

500.0

200.0

Vg=2

Vg=2.5

Vg=2

Vg=2.5

150.0

ID (uA)

400.0

ID (uA)

Vds

0

0.2

0.4

0.6

0.8

1

1.2

1.5

1.7

1.9

2.1

2.3

2.5

2.5

300.0

200.0

100.0

50.0

100.0

0.0

0

0.0

0

0.5

1.5

2.5

0.5

1.5

-50.0

VD (V)

VD (V)

Problem 3

This problem deals with a CMOS inverter with the following parameters: VDD = 3V, Vtn = 0.6V, Vtp = 0.82V, kn = 100A/V2, n = 2.2p.

Solution

a) Determine the beta ratio,

1.3V.

n/ p,

p. 2

2.5

Problem 4

A CMOS inverter with minimum sized transistors has n =0.2mA/V2, p = 0.1mA/V2 and

Vtn=|Vtp|=0.6V.Assume VDD = 3.3V.

a) What is the inverter gate switching threshold (midpoint) voltage VM?

b) What is the resistance for each transistors using our general expression for MOSFET resistance in

saturation?

c) What are the rise and fall times of this circuit if the parasitic capacitance at the output is 9fF?

d) If a load capacitance, CL = 25fF is added to the output, what are the new rise and fall times?

e) What are the propagation delays for this circuit considering both parasitic and load capacitances?

Solution

a) VM =

1+

= 1.47V

1+ 2

1

= 1.85k

n (VDD Vtn ) (0.2)(2.7)

1

1

Rp =

=

= 3.7 k

p (VDD Vtp ) (0.1)(2.7)

b) Rn =

n

p

n

p

d) Here, Cout = 25+9 = 34fF, so

t HL = 2.2 Rn Cout = 2.2(1.85k)(34 fF ) = 138.5 ps

e) t HL = ln(2) Rn Cout = ln(2)(1.85k)(34 fF ) = 43.6 ps

Problem 5

A CMOS inverter is designed with p = 80A/V2, n =0.25mA/V2, Vtn=|Vtp|=0.5V and VDD = 2.5V.

The total capacitance at the output is 50fF

a) Using our general expression for MOSFET resistance in saturation, what is the resistance for each

transistor?

b) What is the rise time of this circuit?

c) What is the fall time of this circuit?

d) What is the high-to-low propagation delay for this inverter?

e) What is the low-to-high propagation delay for this inverter?

f) Do the relative results (rise vs. fall time, high-to-low vs. low-to-high propagation delay) make sense

considering the relative pMOS and nMOS beta values? Briefly explain why or why not.

Solution:

1

= 2.0k

n (VDD Vtn ) (0.25 10 )(2.5 0.5)

1

1

Rp =

=

= 6.25k

3

p (VDD Vtp ) (0.08 10 )(2.5 0.5)

a) Rn =

p. 3

c) t f = 2.2 Rn Cout = 2.2(2k)(50 fF ) = 220 ps

d) t HL = ln(2) Rn Cout = ln(2)(2k)(50 fF ) = 69.3 ps

e) t LH = ln(2) R p C out = ln(2)(6.25k)(50 fF ) = 217 ps

f) Because bn > bp, we should expect the fall time and high-to-low delay to be

lower (faster) than the rise time and low-to-high delay. Calculation results confirm.

Problem 6

A clever ECE410 student has noticed that the simulated worst-case rise time of a CMOS NAND gate is

actually slower than the worst-case fall time, contrary to calculations based on the models give in class.

The student proposes that a more accurate model of the NAND gate worst-case rise time should include

an additional parameter.

a) Specify the input transition sequence that would produce the worst-case rise time. Refer to your results

from Lab 4 if necessary. Format your response like AB CD, where the first number represents the

input to the series transistor that is connected to the output.

b) Sketch an RC model of the NAND gate during this transition including all parasitic capacitances and

resistances for all transistors that are turned on.

c) Derive an improved equation for the worst-case rise time based on your RC model. You should see

that an additional term is necessary to more accurately model this transition.

d) Based on this improved mode, will the worst-case rise time always be longer than the worst-case fall

time? Briefly explain.

Solution

a) The worst-case rise time occurs when 11 10, turning off the

nMOS connected to ground but leaving the nMOS connected to the

output turned on. Only one pMOS is turned on to pull the output

high.

b) The RC model shown here includes the resistance of the nMOS

transistor that remains on during the 1110 transition and is

connected to the output which is changed from low to high during

this input transition.

VDD

Rp

Cout

Rn

Cx

c) For the RC circuit shown here there are two time constants,

RpCout, and (Rp+Rn)Cx. The rise time is thus,

tr = 2.2 (RpCout + (Rp+Rn)Cx)

d) No, it wont always be longer. The rise and fall times are given by

tr = 2.2 (RpCout + (Rp+Rn)Cx)

tf = 2.2 (2RnCout + RnCx)

Depending on the ratio Cout/Cx and the ratio Rp/Rn, the fall time could be longer than the

rise time. For example, if Rp<2Rn and Cout>>Cx the fall time will be longer. Thus, the size

of the transistors and the output load capacitance will determine which output transition

is slowest. This is a feature of the NAND gate, which has very similar rise and fall

characteristics, unlike the NOR gate where there is an obvious distinction and the fall

time is always faster than the rise time.

p. 4

- Transistor Level Implementation of CMOS Com Bi National Logic CircuitsUploaded bySuneel Kushwaha
- sylUploaded byAleksey Pinkhasov
- Verilog codesUploaded bykavi_mishra92
- Datasheet(1)Uploaded byMartin Recman
- etdUploaded byAmar Verma
- Vlsi Lab Report FinalUploaded byvenki
- Ultra-Low Power Digital Subthreshold Logic CircuitsUploaded byIrfan Ahmad Pindoo
- Redefining CMOS Logic Style for SubthresholdUploaded byAnupam Gajendra Narayan Choudhary
- Osciladores Ingles 1Uploaded byEdwin Mauricio Ariza
- mosfetUploaded byLian Phạm
- Syllabus Amie Syllabus Section b (EC)Uploaded bysandeep
- Low-Power Level Shifter ForUploaded byvkry007
- TC4469Uploaded byDev Kumar
- VLSI Ecad LAboratoryUploaded byVenkata Ramakrishna Nandyala
- all related questions- with 1 bUploaded byrdxz12
- 372fbc17145f4f0c7c8947c7f2522d4d.pdfUploaded bygipsun
- P Channel MOSFET:Uploaded bykrishna4130
- Ut28f256qle PromUploaded byDeepa Devaraj
- Si_ncUploaded byGerardo Texis
- Juge_NanoTera_2011.pdfUploaded byParvathy S Parvathy
- Semiconductor_Memory_09_SRAM_A_2007-1Uploaded bysreeramiitm
- POR_Circuit_ANUploaded bymanojr_cet
- IRF9530Uploaded byAnderson Silva
- SEECAUploaded bySunil Pillai
- EXTC.pdfUploaded byAbhishek Shrikant Jangam
- Analog CircuitsUploaded byLubab Salman
- Ngspice ManualUploaded byLalan Joshi
- PhysRevLett.45.494Uploaded byPavel Castro-Villarreal
- Ngspice ManualUploaded byRommel Joshua Ortega
- KS0107bUploaded byrichard70

- WINSEM2015-16_CP1432_28-Jan-2016_RM01_Flip-flopsUploaded bymurthy
- MotivationUploaded bymurthy
- Test.txtUploaded bymurthy
- question description.txtUploaded bymurthy
- Webservices FinalUploaded bymurthy
- Block DiagramUploaded bymurthy
- New Text Document (3).txtUploaded bymurthy
- 1.txtUploaded bymurthy
- CheckUploaded bymurthy
- EmotionUploaded bymurthy
- FALLSEM2016-17_6847_RM001_11-JUL-2016_HUM101_THUploaded bymurthy
- Cheque TemplateUploaded bymurthy
- Cheque TemplateUploaded bymurthy
- template.txtUploaded bymurthy
- chekc_1Uploaded bymurthy
- FALLSEM2016-17_ECE401_ETH_1566_29-SEP-2016_RM001_4Uploaded bymurthy
- FALLSEM2016-17 ECE401 ETH 1593 09-AUG-2016 RM001 Link Impairment Consideration and System DesignUploaded bymurthy
- FALLSEM2016-17 ECE403 TH 1613 02-SEP-2016 RM001 Delay Spread and Time Dispersion ParameterUploaded bymurthy
- FALLSEM2016-17_ECE401_ETH_1593_21-JUL-2016_RM002_Types of FiberUploaded bymurthy
- Fallsem2016-17 Ece401 Eth 1593 26-Sep-2016 Rm001 Non Linear EffectsUploaded bymurthy
- GMSKUploaded bySantosh Ban
- FALLSEM2016-17_ECE403_TH_1613_07-SEP-2016_RM001_coherence BW Doppler Spread and Coherence Time Different Fading Scenarios-ilovepdf-compressedUploaded bymurthy
- Sonet Sdh-protectionUploaded byVamsi Krishna
- FALLSEM2016-17_ECE401_ETH_1566_19-SEP-2016_RM001_3Uploaded byShivendra Rajput
- The Seven Habits of Highly Effective PeopleUploaded byAnkit Verma
- FALLSEM2016-17_1566_RM001_08-AUG-2016_ECE401_ETHUploaded bymurthy
- FALLSEM2016-17_ECE401_ETH_1566_01-NOV-2016_RM001_5Uploaded bymurthy
- FALLSEM2016-17_ECE401_ETH_1566_18-AUG-2016_RM001_2Uploaded bymurthy
- Fallsem2016-17 Ece401 Eth 1593 11-Jul-2016 Rm001 Introduction-session-1Uploaded bymurthy
- gmskUploaded byHuỳnh Công Vị

- EE447_lecture2Uploaded bychakkaravi
- 1.Aging-Aware Reliable Multiplier Design With Adaptive Hold LogicUploaded byPidikiti Surendra Babu
- digital integrated circuit analysis and design solution manual1 pdf.pdfUploaded byAmrYassin
- Kupdf.net Pucknell Vlsi Design PDFUploaded byVishav Verma
- 4-45nm SRAM Technology Development and Technology Lead VehicleUploaded byRohit Bhelkar
- lab1Uploaded bybella_duane
- Transmission GateUploaded byradeonleon
- passgateUploaded byBruno Silva
- CircuiTikz ManualUploaded byKonstantinos Mamis
- Logic Style 1Uploaded byRodrigoMansueliNunes
- Two Subthreshold Full Adder Cells in 65 nm CMOS TechnologyUploaded byJournal of Computing
- Effects of on Resistance RON to an Analog SwitchUploaded byMai Văn Công
- L030-ICTechnology(2UP)Uploaded byMustafaMuhammad
- Solution Gates 2 SamplesUploaded byluqmansulyman
- AOI-22Uploaded byBoyoyon Oyense
- AN-6069.pdfUploaded byprasanna
- Lecture_5Uploaded byAnonymous PaUyyhf1OT
- 05735442Uploaded bykothari_rubin
- A High-Voltage Output Driver in a 2_5V 0_25um CMOS Technology.pdfUploaded byoht993
- CMOSunit3+5notes_Premananda_PUNITHGOWDA M BUploaded byPunith Gowda M B
- cvsl(1).pdfUploaded byAnonymous hsb1qnt3B
- IN-GATE'14Uploaded byChintu Gurbani
- Homework 1Uploaded byShahin Shah
- Seminar_GR 12_Ultra-Low Power VLSI Circuit Design for Energy Efficient SystemsUploaded byanon_241362175
- Design and optimization techniques of high-speed VLSI circuitsUploaded bysandeepmahankali
- introduction-to-novel-approaches-to-low-leakage-and-area-efficientUploaded byIJSTR Research Publication
- Inverter DynamicUploaded byMuneer Mattanur
- A Survey of Low Power Wallace and Dadda Multipliers Using Different Logic Full AddersUploaded byesatjournals
- Design of ALU Circuits using Pass Transistor LogicUploaded byVivek Guntumudugu
- 73040644 VLSI Interview Questions With AnswersUploaded byMohamed