DOC PREVIEW
MIT 6 012 - Lecture Notes

This preview shows page 1-2-3-4-5-6 out of 17 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Lecture 10 MOSFET (III) MOSFET Equivalent Circuit Models Outline • Low-frequency small-signal equivalent circuit model • High-frequency small-signal equivalent circuit model Reading Assignment: Howe and Sodini; Chapter 4, Sections 4.5-4.6 6.012 Spring 2009 Lecture 10 1Large Signal Model for NMOS Transistor • Cut-off • Linear / Triode: • Saturation Regimes of operation: Effect of back bias I D = 0 I D = W L μnCox VGS  VDS 2  VT    • VDS ID = IDsat = W 2L μnCox VGS  VT[ ]2 • 1 + VDS[ ] VT(VBS ) = VTo + 2p  VBS 2p[ ] VDS ID VGS VGS=VT VDSsat=VGS-VT 0 0 linear saturation cutoff VGS VBS VDS �ID 6.012 Spring 2009 Lecture 10 2Small-signal device modeling • Small-signal is small –  response of non-linear components becomes linear • Since response is linear, lots of linear circuit techniques such as superposition can be used to determine the circuit response. • Notation: iD = ID + id ---Total = DC + Small Signal Key Points: In many applications, we are only interested in the response of the device to a sm all-signal applied on top of a bias. VGS VBS VDS �ID+id vgs vbs vds + -++ --6.012 Spring 2009 Lecture 10 3Mathematically: With id linear on small-signal drives: iD(VGS, VDS, VBS; vgs, vds , vbs )  ID VGS, VDS, VBS( ) + id (vgs, vds, vbs ) id = gmvgs + govds + gmbvbs Define: gm  transconductance [S] go  output or drain conductance [S] gmb  backgate transconductance [S] Approach to computing gm, go, and gmb. gm  iD vGS Q go  iD vDS Q gmb  iD vBS Q Q  [vGS = VGS, vDS = VDS, vBS = VBS] 6.012 Spring 2009 Lecture 10 4Transconductance In saturation regime: iD = W 2L μnCox vGS  VT[ ]2 • 1 + VDS[ ] Then (neglecting channel length modulation) the transconductance is: gm = iD vGS Q W L μnCox VGS  VT( ) Rewrite in terms of ID: gm = 2 W L μnCox ID ID gm 0 0 saturation 6.012 Spring 2009 Lecture 10 5Transconductance (contd.) Equivalent circuit model representation of gm: G S D B + -vgs gmvgs id 1 100 200 300 400 iD� (μA) vDS (V) VGS + vgs Q ID + id VDS 5432 ID id id = gmvgs VGS 6.012 Spring 2009 Lecture 10 6Output conductance In saturation regime: W 2iD = 2L μnCox [vGS  VT ]•[1 + vDS ] Then: W 2 iDgo = = μnCox (VGS  VT )• ID2L vDS Q Output resistance is the inverse of output conductance: 1 1 r = = o go ID Remember also: 1  L Hence: r  Lo 6.012 Spring 2009 Lecture 10 7Output conductance (contd.) Equivalent circuit model representation of go: G S D B + -vgs ro id iD� (μA) ID + id vds VDS VDS + vds VDS (V) ID id id = govds VGS, VBS 1 100 200 300 400 Q 5432 6.012 Spring 2009 Lecture 10 8Backgate transconductance In saturation regime (neglect channel length modulation): iD  W 2L μnCox vGS  VT[ ] 2 Then: gmb = iD vBS Q =  W L μnCox VGS  VT( ) • VT vBS Q      Since: Then : VT vBS Q =  2 2p  VBS Hence: VT(vBS ) = VTo + 2p  vBS 2p[ ] gmb = gm 2 2p  VBS 6.012 Spring 2009 Lecture 10 9Backgate transconductance (contd.) Equivalent circuit representation of gmb: G S D B + -vgs gmbvbs + vbs -id iD� (μA) ID + id VDS VDS (V) ID id id = gmbvbs VGS, VBS + vbs VGS, VBS 1 100 200 300 400 Q 5432 6.012 Spring 2009 Lecture 10 106.012 Spring 2009 Lecture 10 Metal interconnectto gateMetal interconnect to bulkn+ polysilicon gatexy0p-typeQN(y)Xd(y)+-+- VDS VGS VBS = 0n+ source n+ drainFigure by MIT OpenCourseWare.6High-frequency small-signal equivalent circuit model Need to add capacitances. In saturation: fringe electric field lines \ gate......................................................source ............................................................................................................................................................................................................................................................................................................................................................... , n+ =',b 4k ~N(%s) 4k1%depletion] overlapL-overlap LD region I Cgs = channel charge + overlap capacitance, C,, Cg, =overlap capacitance, C,, CSb= sourcejunction depletion capacitance (+sidewall) C, =drain junction depletion capacitance (+sidewall) ONLY Channel Charge Capacitance is intrinsic to device operation. All others are parasitic. 6.012Spring 2009 Lecture 10Inversion layer charge in saturation qN (vGS ) = WQN (y)dy 0 L  = W QN (vC )• dy dvC0 vGS VT  • dvC Note that qN is total inversion charge in the channel & vC(y) is the channel voltage. But: dvC dy =  iD W μnQN (vC ) Then: qN (vGS ) =  W 2μn iD • QN (vC )[ ]2 0 VGS VT  • dvC Remember: QN (vC ) = Cox vGS  vC (y)  VT[ ] Then: qN (vGS ) =  W 2 μnC2 ox iD • vGS  vC (y)  VT[ ]2 0 vGS VT  • dvC 6.012 Spring 2009 Lecture 10 13Inversion layer charge in saturation (contd.) qN (vGS ) =  2 3 WLCox vGS  VT( ) Gate charge: qG (vGS ) = qN (vGS )  QB,max Intrinsic gate-to-source capacitance: Cgs, i = dqG dvGS = 2 3 WLCox Must add overlap capacitance: Gate-to-drain capacitance — only overlap capacitance: Cgd = WCov Cgs = 2 3 WLCox + WCov Do integral, substitute iD in saturation and get: 6.012 Spring 2009 Lecture 10 14ther capacitances ............................................................ Source-to-Bulk capacitance: csb = WLdvf cj +(2~di&f+w)cjsw where C :Bottom Wall at Vm (F 1em2) Side Wall at VSB(F1cm)Cjsw Drain-to-Bulk capacitance: Cdb =WLdiffj + (2~dij-f+w)cjsw where Cj :Bottom Wall at VDB(FI cm2) Side Wall at VDB(F1cm)Cjsw Gate-to-Bulk capacitance: 6.012Spring 2009 Lecture 10 .WWhat did we learn today? Summary of Key Concepts gm  W L ID ro  L ID Cgs  WLCox High-frequency small-signal equivalent circuit model of MOSFET In saturation: G S D B + -vgs Cgs Cgb Cgd Cdb Csb gmvgs gmbvbs ro + vbs -id 6.012 Spring 2009 Lecture 10 16MIT OpenCourseWarehttp://ocw.mit.edu 6.012 Microelectronic Devices and Circuits Spring 2009 For information about citing these materials or our Terms of Use, visit:


View Full Document

MIT 6 012 - Lecture Notes

Documents in this Course
Quiz #2

Quiz #2

11 pages

Quiz 1

Quiz 1

11 pages

Exam 2

Exam 2

11 pages

Quiz 1

Quiz 1

13 pages

Quiz 2

Quiz 2

14 pages

Load more
Download Lecture Notes
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lecture Notes and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lecture Notes 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?