Data Converters Design And Testing

33 %
67 %
Information about Data Converters Design And Testing

Published on January 6, 2009

Author: syella

Source: slideshare.net

Description

Presentation includes design and testing of analog and mixed-signal integrated circuits such as an operational amplifier, 12-bit charge scaling architecture based digital-to-analog converter (DAC), 12-bit recycling architecture based analog-to-digital converter (ADC) and operational amplifier with floating gate inputs with on-chip BICS

VLSI Design Lab Quiescent Current Testing of CMOS Data Converters Research Presentation By Siva Yellampalli Date: November 5, 2008 (2:00 – 4:00 PM), Fall 2008 Room: EE117 Department of Electrical and Computer Engineering Louisiana State University

VLSI Design Lab Outline of the Presentation IDDQ Testing methodology Testing data converter circuits Physical faults d fault injection technique Ph i l f lt and f lt i j ti t h i Future trends of IDDQ testing Circuit under test BICS for IDDQ testing BICS for ΔIDDQ testing Combined IDDQ, IDDT and oscillation test methodology methodology. IDDT current testing Oscillation based testing Conclusion C li

VLSI Design Lab Outline of the Presentation IDDQ Testing Methodology Need for Testing Data Converter Circuits Physical Faults and Fault Injection Technique IDDQ BICS Design Future trends of IDDQ testing ΔIDDQ BICS Testing Combined IDDQ, IDDT and Oscillation Test Methodology IDDT Current Testing Oscillation Testing Conclusion

VLSI Design Lab IDDQ Testing Methodology VDD VDD IDDQ=0 Q3 Q1 Vout V Vin Q2 Q4 With fault Without fault

VLSI Design Lab Block Diagram of IDDQ Testing VDD PMOS BLOCK INPUTS OUTPUT NMOS BLOCK CUT PASS/FAIL BICS VSS

VLSI Design Lab Need for Testing of Data Converter Circuits Photolithographic process Particle related defects have been observed as major cause of defects in mixed signal circuits. fd f t i i di l i it Physical failures -bridges -opens and -gate oxide shorts (GOS) gate (GOS).

VLSI Design Lab Bridging Faults VDD Bridging faults -short circuit faults in integrated circuits. it t d i it -can appear either at the logical output of a g g p gate or at V0 the transistor nodes internal VA to a gate. Bridge 1: Drain-gate VB Bridge 3 : Drain-source Bridge 2 : Gate-source Gate source

VLSI Design Lab Bridging Faults B id i F l Different bridging faults*: (a) shorting of seven metal lines caused by unexposed photo resist, (b) shorting of four metal lines by solid state particle on the metal mask, (c) shorts and breaks of metal lines caused by scratch in the photo resist, (d) short among multiple metal lines by a metallization defect of 1um in size, (e) short between two aluminum lines due to metallization defects, and (f) inter layer short between two aluminum interconnects in 0 5 um technology 0.5 technology. *R. Rajsuman, “IDDQ testing for CMOS VLSI,” Proc. of the IEEE, vol. 88, no. 4, 2000, pp. 544-566.

VLSI Design Lab Gate Oxide Faults Gate oxide shorts -occur frequently in CMOS technology. -reasons are breakdown of the gate oxide and th manufacturing t id d the fti spot defects in lithography and processes on the active area and polysilicon masks masks. Gate Oxide defects: (a) gate- oxide short to N+ diffusion*, and (b) gate oxide pin hole causing cell word line leakage memory+ . *C.F. Hawkins and J.M. Soden, “Reliability and electrical properties of gate oxide shorts in CMOS ICs,” Proc. Int. Test Conf., 1986, pp. 443- 451. +A. Chan, D. Lam, W. Tan and S.Y. Khim, “Electrical failure analysis in high density DRAMs,” Proc. Int. Test Conf., 1998, pp. 43-52.

VLSI Design Lab Open Faults (a) a foreign particle causing a line open and a ling thinning* (b) A contaminated particle causing 7- 7 line opens* and (c) SEM picture of defect which caused open in metal 2 and a short p in metal 1+. *W. Maly, “Realistic fault modeling for VLSI testing,” IEEE Design Auto. Conf., 1987, pp. 173-180. +J. Khare, S. Griep, H.D. Oberle, W.Maly, D.S. Landsiedel, U. Kollmer and D. M. H. Walker, “Key attributes of an SRAM testing strategy required for effective process monitoring,” IEEE Int. workshop on Memory Testing, 1993, pp. 84-89.

VLSI Design Lab Fault Injection Technique Example: Fault-Injection Transistor (N- MOSFET) at Inverter Output VDD VE G ME VI V0 S D D (4.5/1.6) (4.5/1.6) ME VE G S FIT VSS

VLSI Design Lab Outline of the Presentation IDDQ Testing methodology Need for Testing data converter circuits Physical faults d fault injection technique Ph i l f lt and f lt i j ti t h i IDDQBICS design Future trends of IDDQ testing g BICS for ΔIDDQ testing IPS current testing Oscillation based testing Conclusion

VLSI Design Lab Outline of the Presentation IDDQ Testing Methodology Need for Testing Data Converter Circuits Physical Faults and Fault Injection Technique IDDQ BICS Design Future trends of IDDQ testing ΔIDDQ BICS Testing Combined IDDQ, IDDT and Oscillation Test Methodology IDDT Current Testing Oscillation Testing Conclusion

VLSI Design Lab VDD(+2.5V) XFIT 3 M8 M5 VE3 XFIT6 XFIT 2 M7 VE6 VSS VSS VE2 VSS CMOS Amplifier (CUT) VOUT M1 M2 VE5 XFIT 5 V+ - V VBIAS M11 (XFIT8) VSS CC M10 XFIT 1 M6 VSS M3 M4 VE7 VE4 VE1 XFIT 7 XFIT 4 VSS(-2.5V) VDD BICS VDD 5.6/1.6 M5 11.6/1.6 M6 VDD VSS IDEF IREF EXT VSS 6.4/1.6 IDEF - IREF OUTPUT IREF 36/1.6 ( PASS/ FAIL ) M4 92/1.6 3.2/1.6 M1 M3 M2 M0 100/1.6 VENABLE 400/1.6 VSS 36/1.6 VSS(-2.5V)

VLSI Design Lab Simulated Si l t d BICS O t t and M Output d Measured IDDQ f d for Defects

VLSI Design Lab Layout of Built-in Current Sensor

VLSI Design Lab BICS Showing PASS/FAIL Output from HP1660CS Logic Analyzer Corresponding to Fault M5DSS

VLSI Design Lab Future Trends of IDDQ Testing The theoretical basis of IDDQ testing is based upon estimation of defect-free defect free current in the circuit and then setting a limit (threshold) above which a circuit is considered defective. Methods for tti threshold limit M th d f setting th h ld li it - 1μA is considered defect free any number from as low as 10 μA to 100 μA is considered as threshold limit. - Due to large number of devices, the distribution of the measured current is expected to be Guassian. Due to statistical variations IC’s up to mean +3σ are considered variations, IC s defect free. A limit higher than mean +3σ is assumed, above which IC’s are considered defective.

VLSI Design Lab Future Trends of IDDQ Testing When the density of the defect free and defective current defect-free are separated from each other, clear distinction between the good and the defective ICs can be made. In submicron CMOS technology the mean of the distribution of defect free current increases and approaches the IDDQ threshold limit (set from earlier technology) Changing the threshold limit to higher number does not resolve the issue because with high leakage current the change in defect-free and defective current is miniscule and unidentifiable.

VLSI Design Lab Future Trends of IDDQ Testing Sub threshold leakage current. ⎛ Vgs −Vth ⎞ ⎛ ⎛ Vds ⎞ ⎞ ⎜ ⎟⎜ ⎜ ⎜ v ⎟⎟ ⎜ S / ln10 ⎟ ⎟ = k exp⎝ ⎠ ⎜1 − exp⎝ t ⎠ ⎟ I sub ⎜ ⎟ ⎜ ⎟ ⎝ ⎠

VLSI Design Lab Future Trends of IDDQ T ti Ft Tdf Testing Design Methods D i M th d - Reduced temperature - Substrate bias - Switched source impedance leakage reduction technique - Dual threshold method Testing methods - ΔIDDQ Testing - Power supply current testing - Oscillation testing method

VLSI Design Lab Future Trends of IDDQ Testing ΔIDDQ Testing of CMOS ICs - relative methods of the offstate current is taken instead of absolute ones for fault detection. Power supply current based testing method. pp y g -variation of AC ripple in the supply current under the application of AC input stimulus is used for fault detection. Oscillation based t ti method. O ill ti b d testing th d - Converting the CUT is into an oscillator using feedback. - Output oscillation frequency is monitored for faults in the CUT. - Needs no test signal generation, uses simple measurement and has high fault coverage.

VLSI Design Lab Outline of the Presentation IDDQ Testing methodology Testing data converter circuits Physical faults d fault injection technique Ph i l f lt and f lt i j ti t h i Future trends of IDDQ testing Circuit under test BICS for IDDQ testing BICS for ΔIDDQ testing IDDT current testing Oscillation based testing Conclusion

VLSI Design Lab Outline of the Presentation IDDQ Testing Methodology Need for Testing Data Converter Circuits Physical Faults and Fault Injection Technique IDDQ BICS Design Future trends of IDDQ testing ΔIDDQ BICS Testing Combined IDDQ, IDDT and Oscillation Test Methodology IDDT Current Testing Oscillation Testing Conclusion

VLSI Design Lab ∆IDDQ BICS Design and Testing VDD TG1 VTG Discharging input to comparator 4-Bit Binary Discharging Comp. Synchronous y Input to output 4-Bit Comp Counter Comparator pulse Output Output TG2 Count 1 Pulse A 1 B Analog g I/P Comp 0 C C CUT 0 D I DDQ CLK I/P Vref

VLSI Design Lab Layout of BICS

VLSI Design Lab Comparator Design VDD = 2.5V 3/0.6 3/0.6 M8 M10 M12 6.6/0.6 6.6/0.6 5.4/0.6 M3 M4 V OUT N2 N N1 Inv1 0.9/0.6 0 9/0 6 0.9/0.6 0 9/0 6 V NEG V POS M1 M2 30/0.6 30/0.6 M5 M6 M9 M11 M13 3.9/0.6 1.5/0.6 1.5/0.6 12/0.6 M7 GND

VLSI Design Lab Analog-to-Digital Converter

VLSI Design Lab Analog-to-Digital Converter

VLSI Design Lab 3-BIT Flash Architecture ADC 2.5v 3 BIT 3-BIT flash architecture 1.5R based ADC. Designed for operation at R 2.5 2 5 V in 0 5 μm n well 0.5 n-well R CMOS technology. Output Digital The various blocks are word R N 2 -1 to N encoder - voltage reference R - resister ladder - seven comparators R - 8:3 encoder R 0.5R

VLSI Design Lab 3-BIT Charge Scaling Architecture DAC Designed for operation at 2.5 V. VOUT Building Blocks: g 2pF 1pF 0.5pF p Reset R 0.5pF 05 F - Voltage Reference - Binary Switches - Scaling Network D2 D1 D0 -O Operational Amplifier ti l A lifi VReference

VLSI Design Lab Low Voltage Operational Amplifier Design Double Ended Bulk Driving Second Biasing Differential to Single Ended High Swing Stage Circuitry Circuitry Input Stage Conversion Stage Circuitry (Class A Amplifier) VDD VON + VT M3 M17 M21 VON 2VON +VT M15 M4 M14 M19 M12 M23 VON +VT M11 M1 M2 Mb VPOS M6 VNEG M7 VOUT MR Cc M9 M5 M8 M22 M20 M10 M13 M16 M18 Note: For the P-MOSFETS the substrate is connected to VDD

VLSI Design Lab Layout of Op-Amp

VLSI Design Lab 12 bit 12-bit ADC with Induced Faults

VLSI Design Lab Simulation of the Comparator With no Fault

VLSI Design Lab Chip layout of 12 bit ADC in 40 pin 12-bit 40-pin Padframe

VLSI Design Lab Microphotograph of Fabricated Chip Part of 12-bit ADC ΔIDDQ BICS

VLSI Design Lab Experimental Results

VLSI Design Lab Simulated Count Values from SPICE and Experimental Values for all the Five Faults

VLSI Design Lab Equivalent Faults VDD FIT FIT M17 M19 R R OUTPUT INPUT FIT FIT M18 M20 R R VSS VSUBSTRATE Low Voltage op-amp.

VLSI Design Lab Double Ended Bulk Driving Second Biasing Differential to Single Ended High Swing Stage Circuitry Circuitry Input Stage Conversion Stage Circuitry (Class A Amplifier) VDD VON + VT M3 M17 M21 VON 2VON +VT M15 M4 M14 M19 M12 M23 VON +VT M11 M1 M2 Mb VPOS M6 VNEG M7 VOUT MR Cc M9 M5 M8 M22 M20 M10 M13 M16 M18 Note: For the P-MOSFETS the substrate is connected to VDD

VLSI Design Lab Scan Path Test

VLSI Design Lab VCO Output ∆IDDQ BICS Design and Testing

VLSI Design Lab Comparator Design

VLSI Design Lab 12-bit Digital-to Analog Converter Unity Gain y Capacitor Array Architecture Storage TG Switch Buffer Unity Gain Capacitor OPAMP 64 c MSB Array LSB Array 63 Vo CH C C 2C 4C 8C 16C 32C C 2C 4C 8C 16C 32C Sample and Hold Input Circuitry To Generate Digital Input Word Output of Multiplexer VREF Control Signal

VLSI Design Lab Layout of a 12 Bit DAC 12-Bit

VLSI Design Lab

VLSI Design Lab Deviation (%) in Frequency Output of BICS Under Fault Injection Conditions

VLSI Design Lab Outline of the Presentation IDDQ Testing methodology Testing data converter circuits Physical faults and fault injection technique Future trends of IDDQ testing Circuit under test BICS for IDDQ testing BICS for ΔIDDQ testing Combined IDDQ, IDDT and oscillation test methodology. methodology IDDT current testing Oscillation based testing Conclusion C li

VLSI Design Lab Outline of the Presentation IDDQ Testing Methodology Need for Testing Data Converter Circuits Physical Faults and Fault Injection Technique IDDQ BICS Design Future trends of IDDQ testing ΔIDDQ BICS Testing Combined IDDQ, IDDT and Oscillation Test Methodology IDDT Current Testing Oscillation Testing Conclusion

VLSI Design Lab Test methodology VDD CUT Input Functional Output Pass/Fail fOSC Additional BICS Feedback Circuitry

VLSI Design Lab IDDT based testing methodology AC ripple in the p pp power supply current IDDT, passing pp y p g through VDD under the application of an AC input stimulus is measured. The i Th input signal should produce a noticeable amount of ti l h ld d ti bl tf difference between the IDDT of each faulty and fault-free cases hence a square wave is used. Tolerance limit for the magnitude of IDDT is set as ±5% such that it takes in account the deviations of significant technology and design parameters parameters. If the magnitude of IDDT with fault falls out of this tolerance range the fault is detected. g

VLSI Design Lab IDDT based testing methodology A small resistor is used to sense th voltage corresponding the lt di VDD to IDDT. This resistor shows 100 Ω VR insignificant variation on CUT i i ifi t i ti CUTs performance. Input signal to the CUT is 4 V p g PMOS p-p, 5 kHz or 1 kHz square kH kH BLOCK wave. INPUT OUTPUT NMOS BLOCK CUT Block diagram of power supply current, IDDT based testing VSS

VLSI Design Lab Two stage CMOS op-amp VDD(+2.5V) XFIT 3 M5 VE3 XFIT 2 XFIT 6 M7 M8 VE2 VE6 VSS VSS VSS VOUT VE5 XFIT 5 V+ V- M2 M1 VBIAS M11 CC VSS M6 VSS XFIT 1 M10 VE7 VE4 M4 VE1 M3 XFIT 7 XFIT 4 VSS(-2.5V)

VLSI Design Lab Spice Simulated IPS for CUT 1

VLSI Design Lab Measured IDDT (AC Ripple) for CUT1 ( pp ) For No Fault, Faults 4 and 6 are Input Voltage injected one at a time for CUT Input is a square wave 4 V p-p at 5 Voltage kHz. kHz Corresponding to IPS IDDT is 74 µA sensed across a 100Ω resistor

VLSI Design Lab SPICE Si l t d and E p i nt l R lt f Simulated nd Experimental Results for CUT 1 IPS in µA IPS in µA Fault Number (Simulated) (Si l t d) (Experimental) (E i t l) No fault 74 74 Fault F lt 1 804 830 Fault 2 95 130 Fault 3 93 120 Fault 4 74 74 Fault 5 170 - Fault 6 73 74 Fault 7 156 268 Loss of IPS. +

VLSI Design Lab Oscillation based test method Consists of converting the R1 R2 CUT into an oscillator using passive/ active components in the fOSC feedback. C R

VLSI Design Lab Two stage CMOS op-amp VDD(+2.5V) XFIT 3 M5 VE3 XFIT 2 XFIT 6 M7 M8 VE2 VE6 VSS VSS VSS VOUT VE5 XFIT 5 V+ V- M2 M1 VBIAS M11 CC VSS M6 VSS XFIT 1 M10 VE7 VE4 M4 VE1 M3 XFIT 7 XFIT 4 VSS(-2.5V)

VLSI Design Lab FFT Simulations for Evaluating the Oscillation Frequency of the CUT Oscillator 2 1 V o ltag e, (V ) 0 ( 1.5E-05 2.0E-05 2.5E-05 3.0E-05 3.5E-05 -1 -2 -3 T ime (s) The natural oscillation frequency (f NAT) was observed to be 875 kHz.

VLSI Design Lab Monte-Carlo Analysis Nominal N i l range of oscillation f f ill ti frequencies i d t i is determined using M t i d i Monte- Carlo analysis by including the tolerance of components and parameters of the CUT. A device tolerance of 5% has been used for the frequency varying components R1, R2, R, C. Process variation effects on the electrical characteristics (viz., Vth, µ0 etc.) need to be considered while evaluating the fault tolerance. A lot tolerance 5% for Vth has been considered here. It can been simulated using SPICE command .MC |# runs| [Anal Type] |output variable| {list} {output | output specification}.

VLSI Design Lab Tolerance band of Oscillation Frequencies With 5% tolerances for the components (R, C, R1, R2) and with 5% for Vth , the tolerance band of oscillation frequency was observed to be [-3.7%, +4.1%] and f MAX = 910 kHz f MIN = 842 kHz Note: The tolerance band is calculated as follows: [Min, Max] = [(f MIN- f NAT)/ f NAT, (f MAX- f NAT)/ f NAT]

VLSI Design Lab Measured Oscillation Frequency

VLSI Design Lab % Deviation from Natural frequency

VLSI Design Lab Conclusion BICS for IDDQ t ti have been d i f testing h b designed f f lt d for fault detection. Circuits like op-amp, ADC and DAC have been designed op amp, as CUTs. Faults have been introduced using FITs. A new BICS for ΔIDDQ testing has been presented. Testing methods combining IDDQ, power supply current and oscillation current have been implemented implemented.

VLSI Design Lab Thank Th k you

Add a comment

Related presentations

Related pages

Design, Modeling and Testing of Data Converters ...

Titel: Design, Modeling and Testing of Data Converters Autoren/Herausgeber: Paolo Carbone, Sayfe Kiaei, Fang Xu (Hrsg.) Aus der Reihe: Signals and ...
Read more

Design, Modeling and Testing of Data Converters eBook von ...

Lesen Sie Design, Modeling and Testing of Data Converters von mit Kobo. This book presents the a scientific discussion of the state-of-the-art techniques ...
Read more

Data Converter Testing (So Far) - University of California ...

• Data Converters – Data converter testing (continued) • Dynamic tests ... Data Converters- Testing & DAC Design © 2009 Page 2 Data Converter Testing
Read more

Design, Modeling and Testing of Data Converters - Springer

Current and Emerging Trends in the Design of Digital-to-Analog Converters. ... Design, Modeling and Testing of Data Converters Copyright 2014 DOI
Read more

Design Modeling And Testing Of Data Converters

Title: Design Modeling And Testing Of Data Converters Author: Simone Wannemaker Subject: design modeling and testing of data converters Keywords
Read more

Chapter 5 Testing Converters F - Analog Devices

TESTING DATA CONVERTERS 5.1 TESTING DACS 5.1 CHAPTER 5 TESTING DATA CONVERTERS SECTION 5.1: TESTING DACs Walt Kester, Dan Sheingold Static DAC Testing
Read more

Data Converter Testing (So Far) - University of California ...

... Data Converters- Testing & DAC Design © 2010 Page 2 Data Converter Testing (So Far) ... Data Converters- Testing & DAC Design © 2010 Page 27 ...
Read more

Data Converters Design And Testing - Technology

1. VLSI Design Lab Quiescent Current Testing of CMOS Data Converters Research Presentation ...
Read more

Design Modeling And Testing Of Data Converters

Title: Design Modeling And Testing Of Data Converters Subject: Design Modeling And Testing Of Data Converters Keywords: Download or Read Online design ...
Read more