This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: ECE4902 Studio Lab 2 C2009 Carrier Mobility, Channel “on” Resistance Ron MOSFET Threshold Voltage VTHH, VTHp SPICE Parameter Extraction
PURPOSE: The purpose of this lab is to measure the resistive nature of the MOSFET drain—source channel.
Upon completion of this lab you should be able to:
° Recognize that the channel of a MOSFET looks (mostly) resistive for small currents iD
and small values of drain—source voltage vDS.
' Recognize that increasing the gate drive beyond the threshold voltage reduces the channel H H '
on reSIStance Ron, ' Determine the carrier mobility and threshold voltage for the N— and P—channel MOSFETS
in the MC14007 ' Compare the measured data to predicted values based on extracted parameters
Prelab P2—1. In the circuit of Fig. P21, a DVM is used to measure the drain~source resistance Ron as a
function of applied gate—source voltage Vos The following data is obtained: VGS Ron
1.5 V 00
2.0 3180 $2
2.5 710 $2
3.2. 330 $2
4.5 180 S2 Estimate the threshold voltage VTH and unC0x% for this MOSFET (see Writeup section for procedure) Figure P2—1. Lab Exercise: REDUCING ON RESISTANCE Ron
BY INCREASING GATE DRIVE VGS 1.2—]. Use the circuit of Fig. L2—2 (using the DVM in ohmmeter mode) to measure the on
resistance Ron directly. As you vary VGS, you will see variation in the on resistance. Resistor
RG is solely for protection of the MOSFET gate; the voltage drop across RG is zero under
normal conditions since the DC current into the MOSFET gate is zero. NOTE: Use the AFG3021 function generator in DC output mode (access through the button
sequence More > More Waveform Menu > DC). Be sure to set the numerical amplitude
display to the "High Z" load condition so the displayed DC level is correct. As an additional check, use the scope to look at the gate voltage so you can see vGS go up and down as you
adjust the DC level. ADDITIONAL NOTE: It‘s important to observe the DVM lead polarity shown in Fig. 122. If the DVM connection is reversed, the DVM current iDVM will forward bias the substrateto—
drain diode when VGS < VTH. ANOTHER ADDITIONAL NOTE: For this part, it’s very important to leave the DVM on
the 20V/20kQ range throughout. Changing the resistance range changes the current the
DVM uses to measure resistance, which affects the resistance measured in the triode region.
The problem: in the 2k§2 range, the DVM injects a larger test current than in the ZOkQ
range. This causes a larger value of VDS, violating the “small VDs” condition for the resistive
portion of the triode region. So, even though normal practice would be to use the 2kS2 range when measuring Ron less than 2kg (for better measurement resolution), in this case, we’ll
live with lower resolution so as not to violate the “small vDS” condition. L2—2. Set the AFG3021 voltage VDC to provide VGS of approximately +5V. Measure Ron and
VGS. The resistance R0n should be in the range of approximately 2009 to le (which will
read : 0.2kQ to : 1k§2 with the DVM on the ZOkQ range). Reduce VGS until Ron has
increased by about a factor of two. Measure Ron and vGS again. As sz decreases, you
should see R0n increase. Continue decreasing the power supply voltage to decrease vGS in
increments that approximately double the measured value of Ron. Measure Ron and v05 at
each step and make a table of measurements in your lab notebook. As sz gets closer to the
threshold voltage VTH, Ron will start increasing rapidly. As Ron increases, take a few data points with Ron values of about 2kg, 5kg, and 10kg. You should end up with a total of
about 5 to 10 data points. In your lab notebook, plot Ron as a function of VGS . Note that increasing gate drive vGS
reduces the value of onresistance. Also, plot 1/ Ron as a function of VGS. You should see an approximately linear relationship between 1/ Ron and VGS; a line through the measured
data points should intersect the vGS axis at the threshold voltage. iDVM AFCS 3OZ R6 §M1
g1/6 CD4007/ M04007 DVM Figure L2—2. L2—3. Repeat the above procedure for a P—channel MOSFET as shown in Figure 123 below.
Doublecheck your connections with the AFG3021 and the DVM since you will now be
measuring VSG while applying a negative voltage to the gate. The range of resistances Ron should be similar to the previous part; follow the same procedure for gathering data, In your lab notebook, plot Ron as a function of VSG . Also, plot 1 / Ron as a function of v50.
Note that, since your horizontal axis is vSG rather than v65, some care is needed in
interpreting the plots. For the P—channe] MOSFET making the gate drive voltage vGS
more negative reduces the value of onresistance. On the 1/ Ron plot, you should also see
an approximately linear relationship between 1/ Ron and VSG; however in this case a line
through the measured data points should intersect the vSG axis at the negative of the threshold voltage. So, for example, if the vSGintercept is at 2V, then the threshold voltage
is V m : ~2V. Figure L2—3. WRITEUP
RELATIONSHIP BETWEEN ON RESISTANCE rum") AND GATE DRIVE VQS NChannel MOSFET
W2l , Plot Ron as a function of sz. W
W2—2. Plot 1/ Ron as a function of v05. Using this plot, extract the slope unCox—L— , and the x—intercept (threshold voltage VTH ) for the triode region "on" resistance expression: 1 Run = _T“_ (21) Kurt OXT(VGS _VTH) W . .
W2—3. From the slope unCox—L—, determine the value of mobility [4". Use your value of Cox from Lab 1 and the W/L = 350/10 for the MC14007. W24. Using your parameters from W2—2, plot the prediction of the MOSFET Ron model on the same axes with your measured data from part L2—2. How well does the model predict the
measured data? PChannel MOSFET W2—5. Plot Ron as a function of v30. W
W26. Plot 1/ Ron as a function of v50. Using this plot, extract the slope MPC , and the 0x — L
x—intercept (which will be —VTH , the negative of the p—channel threshold voltage) for the
triode region "on" resistance expression: Ron = +_ (21)
Hp 0x $050 " (—VTH )) W2—7. From the slope M pCox Kg, determine the value of mobility up. Use your value of Cox from Lab 1 and the W/L = 900/10 for the MC14007. W2—8. Using your parameters from W2—6, plot the prediction of the MOSFET Ron model on the same axes with your measured data from part L23. How well does the model predict the
measured data? NOTE: The above procedure can be considerably simplified by using the MATLAB file
rdsplot .m, available from the course website. Look at the source code to see where to put in
your measurements. Your plots will look something like Figure L2—4 below. ...
View
Full Document
 Spring '01
 MCNEILL
 Integrated Circuit

Click to edit the document details