Model 4210-CVU Applications Training

Transcription

1 Model 4210-CVU Applications Training Mark Walter Updated May, A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

2 4210-CVU Applications Training Agenda CVU Overview Hardware and Connections Projects and Applications KITE Set-up and Features Measurement Tools and Optimization Troubleshooting Questions? 2 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

3 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y CVU Measurement Overview The CVU is a multi-frequency (1kHz-10MHz) impedance measurement card. Measures impedance by sourcing an ac voltage across the DUT and measuring the resulting ac current and ac voltage. ***one ac voltage measure range: 100 mv rms ***three ac current ranges: 1 A, 30 A, and 1 ma The time domain of the ac current and ac voltage must be processed into the frequency domain to produce the phasor form of the DUT impedance. The capacitance and conductance are derived parameters from the impedance and phase. DC voltage bias can be applied to the DUT (+/- 30V). 3 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

4 Simplified Block Diagram 4210-CVU ABB Feedback Ac Source AC Voltmete r Amme Ac t er Simplified Block Diagram HCUR HPOT LPOT LCUR Measures AC impedance (Z DUT ) of the DUT by sourcing an AC voltage across the device and measuring the AC voltage and the resulting AC current. The time-domain AC values are processed into the frequencydomain to produce the phasor form of the impedance. 4 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

5 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Measured Parameters Z, Theta Impedance and Phase Angle R + jx Resistance and Reactance Cp-Gp Parallel Capacitance and Conductance Cs-Rs Series Capacitance and Resistance Cp-D Parallel Capacitance and DissipationFactor Cs-D Series Capacitance and Dissipation Factor Z Cs-Rs Cp-Gp Z X X R 2 2 Z = R + jx X arctan R = Zcos X = Zsin 1 Y G R jb R Z Where: Z=impedance =phase angle R=resistance X=reactance Y=admittance G=conductance B=susceptance 5 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

6 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Calculating Inductance Even though the 4210-CVU does not measure inductance directly, the inductance can be extracted in the Formulator from the Impedance (Z), Phase Angle (theta, ), and the Test Frequency (f). X = Zsin L = X/2 f The Measured Options Parameters must be set to Z, theta. NOTE: The units for theta are in degrees. To use the trig functions in the Formulator, you must convert to rads using the rad function as shown below. A G R E A T E R M E A S U R E O F C O N F I D E N C E 6 Copyright 2004 Keithley Instruments, Inc.

7 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Accuracy Specs The accuracy is specified at specific capacitance values: 1pF 10pF 100pF 1nF 7 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

8 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Maximum Capacitance The measurement accuracy of the 4210-CVU is specified up to 100 nf. However, the CVU can make much higher capacitance measurements. The maximum capacitance is based on test frequency, ac drive voltage, and range. The Capacitance Range Estimator in the CVU Terminal Properties Window shows CMax values in the mf range. However, capacitance values above 100 nf are not tested or calibrated. In general, to measure high capacitances, use the lowest test frequency (1 khz), the smallest AC drive voltage (10 mv rms), and the maximum current range (1 ma or autorange). 8 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

9 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Minimum Capacitance The Minimum Capacitance that can be measured by the 4210-CVU is a difficult question to answer because there are many variables in the test system such as test frequency, ac drive voltage, cables, device factors, open correction, switch matrix, etc. However, based on the specifications with a 1 MHz test frequency and 30 mv rms drive voltage, the measurement accuracy of a 1 pf capacitor is: +/-.38% The range of measured capacitance: pf to.9962 pf 9 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

10 Measuring Hundreds of Femto-Farads A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright Keithley Instruments, Inc.

11 4210-CVU Applications Training Agenda CVU Overview Hardware and Connections Applications and Projects KITE Set-up and Features Measurement Tools and Optimization Troubleshooting Questions? 11 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

12 Supplied Cables and Adaptors CA-447A SMA Cable SMA Cable SMA Cable SMA Cable CS-1247 SMA to BNC Adaptor SMA to BNC Adaptor SMA to BNC Adaptor SMA to BNC Adaptor CS-701A BNC Tee BNC Tee 4210-CVU is supplied with: (4) CA-447A SMA to SMA 1.5m Cables, (red cables) (4) CS-1247 SMA to BNC Adaptors, and (2) CS-701A BNC Tees Always remember... a) BLACK CABLES = IV b) WHITE CABLES = Pulse c) RED CABLES = CV 12 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

13 4200-CVU-PROBER-KIT The 4200-CVU-Prober-Kit was designed to cover a wide variety of prober and manipulator types. For example, some probers have triax feed-thrus on the rear panel, while others have direct coax connections to the manipulators themselves. This kit contains a combination of triax and coax adaptors and barrels that will handle almost any prober you might have. DON T FORGET TO JUMPER THE SHIELDS TOGETHER AT THE PROBE TIPS! 13 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

14 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Connections to Prober with Triax Connectors To connect to probers with triax connectors: (2 or 4) 7078-TRX-BNC male triax to female BNC adaptor (removes outside shield of triax) -OR- (2 or 4) 7078-TRX-GND male triax to female BNC adaptor (removes inside shield of triax) -OR- (1) 4200-CVU-PROBER-KIT contains both above choices PLUS a variety of SMA, BNC, triax adaptors and 4200-PRB-C (jumper wire to connect shields together near DUT) 14 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

15 4210-CVU Applications Training Agenda CVU Overview Hardware and Connections Applications and Projects KITE Set-up and Features Measurement Tools and Optimization Troubleshooting Questions? 15 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

16 CVU ITMs in Default Project The default project includes ITMs for CV measurements. 1) MOSFET: cv-nmosfet (modified TG-439 MOSFET) 2) Diode: cv-diode (RF-43 diode, 1N3595) 3) Capacitor: cv-cap (10 pf Cap) 16 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

17 cv-nmosfet ITM The cv-nmosfet ITM is for use with the demo DUT. This ITM makes a C-V sweep on a four terminal MOSFET and calculates common parameters such as oxide capacitance, flatband capacitance, and flatband voltage. 17 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

18 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y cv-nmosfet ITM 4210-CVU HCUR HPOT LPOT LCUR G D B The CV measurement is made with the Gate of the transistor connected to the HCUR/HPOT terminals and the Source/Drain/Bulk are tied together to the LCUR/LPOT terminals. S AC Measure I and the DC V Source are both on the HCUR/HPOT terminals. 18 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

19 Setting Up the Demo DUT Demo DUT is a modified TG-439 N-MOSFET which is the standard demo DUT. However, for the CVU demo the Source/Drain/Bulk terminals are soldered together. The modified transistors and the SMA tees will come with the 8101-PIV. 19 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

20 Insert the Demo DUT LCUR/LPOT HCUR/HPOT 20 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

21 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Demo Device Notice the shields of the SMA Cables are connected to the S and B terminals of the 8101-PIV. The Demo DUT is connected between the D and G terminals. To HCUR SMA Cable SMA Cable To LCUR SMA Tee SMA Tee To HPOT To LPOT 21 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

22 Executing the cv-nmosfet ITM 22 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

23 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y cv-nmosfet ITM Added Formulas: C fb V fb C ox T ox R s C adj Doping Added Constants: Area Dopetype Temp ox S 23 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

24 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y CVU Applications and Projects 12 projects have been created to make CV measurements on common devices. These project are in the _CV folder and begin with CVU_. These test modules are complete applications that include formulas for parameter extraction and data from actual devices. 24 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

25 25 A G R E A T E R M E A S U R E O F C O N F I D E N C E The MOScap project has 3 ITMs that extract many of the CV common parameters. This project includes 24 formulas and 9 constants. C opyright 2004 Keithley Instruments, K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y CVU_MOScap Project C-V measurements are used to study the quality of the gate oxide on a MOS structure called a MOScap. (Basically a MOSFET with no Source and Drain.)

26 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Difference Between MOSFET and MOScap CV Curves MOSFET MOScap Accum. Depl. Inver. Accum. Depl. Inver. In the Inversion Region, the capacitance of the MOSFET is high, unlike the MOS cap which has low capacitance in inversion. This is because the MOSFET has a source and drain, which enable inversion charge to flow, unlike the MOS cap which relies on generation and recombination in the bulk region. 26 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

27 Projects and Applications CVU_MOScap ITMs C-2vsV_MOScap ITM performs a 1/C^2 curve. The Doping is related to the reciprocal of this curve. DopingProfile_MOScap ITM displays the substrate doping concentration as a function of depletion depth. 27 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

28 CVU_PVcell Project This project combines both IV and CV tests of a solar cell. The Forward and Reverse Bias I-V measurements are made with the 4200-SMU. C-V and C-f sweeps are generated with the CVU. The device must must have small-sized area. The larger the area, the higher the current and capacitance will be. 28 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

29 CVU_PVcell Project I-V C-V P MAX, V OC, I SC, I MAX, V MAX Doping concentration, built-in voltage A 29 G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

30 CVU_HighV project This user module uses the SMU to source voltage and the CVU to measure the capacitance. This project required the Model CVU-PWR package. C-V from +/- 200V or 400V differential can be made using the CVU, SMU, and bias tees. 30 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

31 4210-CVU Applications Training Agenda CVU Overview Hardware and Connections Applications and Projects KITE Set-up and Features Measurement Tools and Optimization Troubleshooting Questions? 31 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

32 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Force/Function Window Just like an SMU, in this Window is where the test is set-up. Choose a Force Function and set up the voltage and frequency. Choose the measurement parameters. Most devices will be Cp- Gp. This is the default. If change measure parameters all formulas and data will be lost! Notice the orientation of the voltage drop. This was chosen on Definition Page and the Advanced Tab. 32 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

33 Advanced Tab CVU Terminal Properties 4200-CVU Terminals (Connections on rear panel of 4200) AC DC HCUR ACV Force DCV HI Force HPOT ACV Sense DCV HI Sense By default, the DC V-Source in on the CVH1 terminals and the AC Ammeter is on CVL1. However, you have the option to change the function of the terminals. LPOT ACV Sense (0V) DCV LO Sense LCUR ACI Measure DCV LO Force (0V) NOTE: CVH1 = HCUR and HPOT connected together CVL1 = LCUR and LPOT connected together 33 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

34 Advanced Tab CVU Terminal Properties In this Window is where the current range is set. Notice the Capacitance Range Estimator. Based on the current range, test frequency, and the AC drive voltage, the maximum capacitance value is calculated. By default the Range is set to Auto. The offset box enables the user to offset the voltage by up to 30V on one terminal. This enables the user to output voltage from 0V to 60V instead of -30V to +30V. 34 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

35 4210-CVU Applications Training Agenda CVU Overview Hardware and Connections Applications and Projects KITE Set-up and Features Measurement Tools and Optimizations Troubleshooting Questions? 35 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

36 Status When status box is checked information is returned in the sheet that provides information about overflow and range. 36 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

37 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Status and Overflow Indication When overflow conditions occur, the data in the sheet changes color and a message is indicated in the upper left hand corner of the graph. 37 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

38 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Confidence Check Confidence Check is a new feature that has been added to enable fast, real time measurements without using an ITM or a UTM! This is a great diagnostic tool to enable a quick check to ensure proper connections to a DUT, An OPEN, or a SHORT. 38 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

39 Confidence Check The Confidence Check is like a digital display. It can be used to measure a DUT, or perform Open and Short check. The CVU settings can also be changed by clicking on the Settings button. 39 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

40 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y ITM Settings That Affect Measurement Time Timing Menu: Speed Mode: Fast, Normal, Quiet Sweep Delay (sweep mode) Interval Time (sampling mode) Hold Time (PreSoak Time) Autorange: time is added to settle at each range only added to first reading taken once the range has changed: 1 A: 400 ms; 30 A: 200 ms; 1 ma: 100 ms Measurement Range: the lower the current measurement range, the more delay time is added to each reading to settle (implies higher impedance) Test Frequency: the higher the frequency, the more time is added to settle the reading (implies higher impedance) 40 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

41 ITM Timing Menu Speeds: Fast, Normal, Quiet, Custom Modes Fast: Fastest time, highest noise. Normal: Most common setting. Allows sufficient settling times for most measurements. Quiet: High accuracy, slower time setting. Allows more time for DC settling and provides longer integration time. Choosing the appropriate Speed Mode involves a speed vs. noise trade-off. Custom: can be programmed for the highest accuracy (with the settling time). A G R E A T E R M E A S U R E O F C O N F I D E N C E 41 Copyright 2004 Keithley Instruments, Inc.

42 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Fast, Normal, Quiet Modes High Impedance DUT 42 Test Device:.53 pf c A al G c R a E p A ; T E T R e M st E A F S re U q R u E e O n F c C y O : N 1 F I M D H E N z C ; 30 uaran ge Co pyright 2004 Keithley Instruments, Inc.

43 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Normal, Fast, Quiet, and Custom Speed Modes High Impedance DUT The capacitance measurements on the.53 pf cal cap were made at 1 MHz test frequency on the 30 ua range. The average reading time and standard deviation for 30 readings taken shown below: Speed Mode Avg. Time per Reading Std. Deviation Normal 48 ms 234E-18 Fast 39 ms 766E-18 Quiet 138 ms 76.8E-18 Custom (10 plc, 3 filter) 1.5 s 20.4E-18 Notice there is not a large difference between the Average Reading Times between the Normal and Fast modes. This is because a fixed settle time is added to each reading on the lowest current range. 43 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

44 Fast, Normal, Quiet Modes 100 nf Cap, Lower Impedance 44 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

45 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Fast, Normal, Quiet Modes Low Impedance DUT The capacitance measurements on the much lower impedance 100 nf cap were made at 100 khz test frequency with auto-range ON. The total test time and standard deviation for each test are as follows: Speed Mode TotalTest Time (100 readings) Std. Deviation Fast s E-12 Normal s E-12 Quiet s E-12 Notice that with the lower impedance DUT (higher current), that the Fast mode is now about twice as fast as the Normal mode. Less settle time is added to each reading than on a lower current range. 45 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

46 Lab 1 Speed vs. Noise Comparisons 46 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

47 Speed vs. Noise Comparisons of Demo DUT A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Ke 47 ithley Instruments, Inc.

48 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Choosing Appropriate Hold and Sweep Delay Times Choosing appropriate hold and sweep delay times is important to many applications. The condition of a device when all internal capacitances are fully charged after applied step voltage is referred to as equilibrium. an If capacitance measurements are made before the device is in equilibrium, inaccurate results may occur. To choose the delay times for a C-V sweep, step an applied voltage using the Sampling Mode, and plot the capacitance as a function of time. Observe the settling time from the graph. Use this time for the Hold Time for the initial applied voltage or for the Sweep Delay Time applied at each step in the sweep. The Sweep Delay Time may not need to be as long as the first step. You will need to experiment. 48 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

49 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Choosing Appropriate Hold and Sweep Delay Times To choose the delay times for a C-V sweep: -step an applied voltage using the Sampling Mode -use initial conditions of Hold Time=0, Interval Time=0, and PreSoak V =0 -plot the capacitance as a function of time. -observe the settling time from the graph. -use this time for the Hold Time for the initial applied voltage or for the Sweep Delay Time applied at each step in the sweep. The Sweep Delay Time may need to experiment. not need to be as long as the first step. The user will 49 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

50 Example: Equilibrium Time Measurement A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright Keithley Instruments, Inc.

51 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Analyzing MOS Cap C-V Curves for Equilibrium C-V measurements on MOS caps should only be made under equilibrium conditions. A MOS cap takes time to charge up after a voltage step is applied. To allow the MOS cap to reach equilibrium: Allow a sufficient Hold Time while applying the PreSoak Voltage. This is the voltage applied prior to the C-V Sweep. After each DC voltage step of the sweep, allow an adequate Delay time before taking the measurement. There are two primary indicators that can be used to determine whether a device has remained in equilibrium. 1) Check the shape of the C-V curve when sweeping from either direction (hysteresis), and 2) The curves should exhibit a smooth, equilibrium shape. 51 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

52 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Analyzing MOS Cap C-V Curves for Equilibrium One way to determine if the device is in equilibrium is to check the shape of the C-V curves in the inversion region when the sweep direction is changed. If the device is settled at all points of the sweep, it should make no difference if the sweep goes from accumulation to inversion or from inversion to accumulation. The dotted lines in the curves indicate unsettled readings in the inversion region. The solid line indicates measurements taken in equilibrium. 52 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

53 C-V Curves of MOS Cap Showing Non-Equilbrium 3) Equilibrium 1) Sweep Delay time too fast 2) Hold Time too short A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Ke 53 ithley Instruments, Inc.

54 Determining the Equilibrium Time of a MOS Cap A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright Keithley Instruments, Inc.

55 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Reducing the Effects of Stray Capacitance Offset Connection Compensation: Open: Corrects for large impedance, small capacitance offsets Short: Corrects for low impedance, high capacitance offsets Guarding: Wafer Level Measurements, Packaged Devices Proper Connections of the Ammeter Terminal of the CVU: Effects of capacitive loading of the terminals on the ammeter 55 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

56 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Open/Short Correction Offset errors are compensated for by using global corrections found in the Tools Menu in KITE. 100 nf 10 nf 1 nf Offset Nominal Short Correction High Capacitance 100 pf 10 pf 1 pf Open Correction Low Capacitance 1pF 10pF 100pF 1nF 10nF 100nF 56 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

57 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y OPEN/SHORT/LOAD/CUSTOM Correction Correction is a two part process. The corrections are performed, and then they are enabled within an ITM. To perform the corrections, Open the Tools Menu and select CVU Connection Compensation. For an Open correction: click on Measure Open. Probes must be up or DUT should be removed from the test fixture. For a Short correction: Connect short between all output terminals. Use BNC barrel or short probes. For a Load correction: Use standard resistance value. Enter value in box. For Custom: Connect up custom cables measure. and 57 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

58 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y OPEN/SHORT/LOAD Correction To enable corrections, click on the Compensation button in the Force/Function Window. Only click on the corrections to be applied. The performed corrections are stored in the 4210-CVU board and once performed can be used in any project. If cabling/connections in the test circuit change, then the corrections will need to be performed again. 58 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

59 Lab 2 Performing Open Correction 59 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

60 C-V Curve Before and After Open Correction on MOSFET Demo DUT A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithle 60 y Instruments, Inc.

61 on C-V 100 Curve Before and After Correction nf Capacitor (low impedance device) A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright Keithley Instruments, Inc.

62 Guarding Guarding can reduce the effects of stray capacitance from affecting measurement accuracy. The Guard terminal is the outside shield of the coax cables of the 4210-CVU. Examples of Guarding: 1. When only one parameter (between two terminals) of a multiterminal device (three or more terminals) is being measured. The unused terminals should be guarded. 2. Guard the chuck for capacitance measurements between terminals on a wafer. 62 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

63 Measuring One Parameter (C BE ) of BJT CVU HCUR Base C BC Collecto r C CE HPOT LPOT C BE LCUR Emitter C C C BC CE MEAS C BE C C BC CE 63 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

64 Guarding Stray Capacitance Guard Connection Collector 4200-CVU HCUR HPOT Base C BC C CE LPOT LCUR C BE Emitter With the Guard Connection, C MEAS = C BE. 64 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

65 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y The Effects of Guarding on Measurement Accuracy Unguarded Guarded A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley In 65 struments, Inc.

66 Lab 3 Guarding a Terminal of a MOSFET 66 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

67 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Guarding the Effects of Stray Capacitance on a MOSFET A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keit 67 hley Instruments, Inc.

68 Guarding Measurements on a Wafer 4210-CVU LCUR LPOT HPOT HCUR Capacitance from DUT Terminal to Chuck C DUT on Wafer Chuck Connect Shields of Cables to Chuck To Guard Stray Capacitance from the Wafer to the Chuck 68 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

69 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Choosing the Optimal Connections By default, the LCUR terminal is the AC Ammeter connection. The HCUR terminal is the DC Voltage Source terminal. However, you have the option to change the function of the terminals in the Advanced Tab CVU Terminals (Connections on rear panel of 4200) AC DC HCUR ACV Force DCV HI Force HPOT ACV Sense DCV HI Sense LPOT ACV Sense (0V) DCV LO Sense LCUR ACI Measure DCV LO Force (0V) NOTE: CVH1 = HCUR and HPOT connected together CVL1 = LCUR and LPOT connected together 69 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

70 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y To avoid noisy measurements and stray capacitance, the AC ammeter terminal should always be connected to the terminal of the device which has the least amount of capacitance to ground CVU CVU HCUR (DC Vsource) HPOT LPOT C DUT LCUR (ammeter) C1, C2 = Parasitic Capacitance to Common C1 Pad 1 C1 < C2 Pad 2 C2 70 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

71 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Example of Improper Connections The AC Ammeter terminal should not be connected to the Pad which has greater capacitance to common! In many cases the user will not know which Pad has the greatest capacitance to ground. However, it is easy to reverse the leads in the advanced tab and repeat the measurement CVU HCUR (DC Vsource) HPOT LPOT ( ammet LCUR er) Example: C1=5000pF C2=5pF Pad 1 C 1 C DUT = Interconnect Capacitance C DUT Chuck Pad 2 C C 2 Metallization Layer Parasitic Capacitance To Common: C 1 C 2 71 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

72 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Results of Proper and Improper Connections Blue Connected correctly (AC Ammeter on Probe.) Purple Improper connections (AC Ammeter on Chuck.) A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instrume 72 nts, Inc.

73 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Connections with Agilent LCR Meter Using the Agilent LCR meter, the only way to correct the connections in this measurement is to physically reverse the leads! When you do this the DC bias is also reversed! Improper Proper Agilent LCR Meter LCUR LPOT HPOT HCUR (AC ammeter) (DC Vsource) Agilent LCR Meter LCUR LPOT HPOT HCUR (AC ammeter) (DC Vsource) C1 > C2 Pad 2 Pad 1 Pad 2 Pad 1 C DUT C 2 C DUT C 2 C 1 C 1 Chuck Chuck A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instr 73 uments, Inc.

74 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y The Importance of Shield Connections Important Rule: Connect the shields of the coax cable together as close as possible to the DUT to avoid errors! This reduces the loop area of the shields which minimizes the inductance. This also helps to maintain the transmission line effects. If the shields are not connected together, large offsets may occur. The higher the frequency the more this becomes important. Connect Shields Together! Shield Shield Cap on Wafer A G R E A T E R M E A S U R E O F C O N F I D E N C E 74 Copyright 2004 Keithley Instruments, Inc.

75 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Cabling Problems Having appropriate cable length is crucial to making successful measurements. Inaccurate results occur with improper cabling. Here are some potential problems with cabling: 1) Improper cable length (mismatched length on different CVU terminals, length compensation incorrect) 2) Improper impedance (not using red cables) 3) Improper shield connections (not connected or not connected close enough to the DUT) 4) Bent, crimped or flattened cables (more critical at higher test frequencies) 5) SMA cable connectors not connected tight enough to test fixture (must be torqued to spec) 75 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

76 4210-CVU Applications Training Agenda CVU Overview Hardware and Connections Applications and Projects KITE Set-up and Features Measurement Tools and Optimization Troubleshooting Questions? 76 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

77 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Troubleshooting Use the Tools available in KITE: Status information Offset corrections, Confidence Check, Ensure proper settings in KITE: drive voltage, dc source voltage, sweep delay times, test frequency, ac etc. Make sure proper cabling: all cabling lengths same, shields connected as close as possible to DUT Ensure verify. good DUT and proper contact to DUT: use Confidence Check to 77 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

78 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y Troubleshooting For making low and high capacitance measurements in general: Low Capacitance: use Open Correction, high test frequency (100kHz, 1MHz), Quiet mode, guard (if applicable), lowest current (or Auto) range, add sufficient sweep delay time for settling. High Capacitance: use Short Correction, low test frequency (1 khz) Higher Frequency: Ensure correct cabling 78 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

79 K E I T H L E Y C O N F I D E N T I A L P R O P R I E T A R Y ABB Unlock Status Error ABB Unlock Errors occur when the ABB is unbalanced. an ABB Unlock Error may occur: Possible reasons 1) Cable length not the same on CVU terminals 2) HPOT or LPOT terminals got disconnected 3) Excessive noise on LPOT terminal 4) Interfering high frequency sources that are measurement frequency the same frequency as the 5) Too high stray capacitance to common Use Confidence Check to help troubleshoot ABB errors! 79 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

80 Troubleshooting Guide Error Symptom Possible Causes Suggestions to Minimize or Avoid Error Capacitance too high Cabling and Connection capacitance affecting measurements Light left ON or Lid Open Perform Offset Correction and Enable in ITM, minimize stray capacitance, minimize cable length if possible Turn off light or close lid Unwanted capacitance from other terminals affecting measurement Shorted DUT Use guarding Try another DUT. Use Confidence Check for verification. 80 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

81 Troubleshooting Guide Error Symptom Possible Causes Suggestions to Minimize or Avoid Error Capacitance too low Device not in equilibrium Increase Delay Time Poor or no contact to device Open DUT Coax cable shields not connected If readings are in the ff range and are expected to be much higher, re-verify connections using Confidence Check. May need to improve contact betweenwafer and chuck. Try another DUT to verify if the problem may be the DUT or something else. Use Confidence Check. Connect shields near DUT. Reduce test frequency. 81 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

82 Troubleshooting Guide Error Symptom Possible Causes Suggestions to Minimize or Avoid Error Noisy Measurements Noisy DUT or environment DUT not shielded electrostatically Use Quiet or Custom Mode; increase or reduce test frequency depending on magnitude of capacitance; verify prober contact to DUT Ensure proper test fixture shielding. The shield must be electrically connected to the coaxial shield. Tails on end of C-V Sweep Device not in equilibrium Set PreSoak voltage to first voltage in sweep and apply sufficient Hold Time to allow DUT to charge up. Leaky Device Try measuring the leakage current using the SMU, reduce the dc voltage 82 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.

83 Conclusion Questions? 83 A G R E A T E R M E A S U R E O F C O N F I D E N C E Copyright 2004 Keithley Instruments, Inc.