EE 462: Laboratory Assignment 5 Biasing N- channel MOSFET Transistor

Transcription

1 EE 46: Laboratory Assignment 5 Biasing N channel MOFET Transistor by r. A.V. adun and r... onohue (/1/07 Udated ring 008 by tehen Maloney eartment of Elecical and Comuter Engineering University of entucky Lexington, Y nsuctional Objectives Analyze the metal oxide semiconductor (MO field effect ansistor (FET (MOFET using a C load line esign a circuit to set a C oerating oint for a MOFET Measure the oerating oints in a C biased FET circuit Become more familiar with rogramming in LabVEW (ee Horenstein 5., 7.3.1, and Background Transistors are nonlinear devices; however, over certain oerating regions they can be aroximated with linear models. To ensure a ansistor oerates in its linear region, a C level is added to its inut signal. The design of this C level is referred to as biasing the ansistor. The C current and voltage values are referred to as the ansistor s C oerating oint (or its bias oint (or its quiescent oint. Once a ansistor is biased in a linear region, small changes for the inut currents and voltages around the bias oint will cause the oututs to change in a linearly roortional manner (aroximately. t is assumed that the variation of the ansistor's currents and voltages are small enough such that they do not move the system into nonlinear oeration regions (iode region or cutoff. The simlest common source MOFET amlifier biasing scheme is shown in Fig. 1. ince only the C oerating oint is of interest right now, the time varying art of the inut signal is omitted. The actual signal would be an AC signal added to the V (Gate to Ground C signal. The ansfer characteristic (outut as a function of the inut for this circuit can be derived. Aly VL in Fig. 1 to obtain: Vout =, (1 V then using a relationshi between and the gate voltage V when the MOFET is in the saturated (forward active region, Eq. (1 becomes: ( Vout = V V V where = (Horrenstein uses and ice uses. (

2 and V is the threshold voltage between the cutoff and iode oeration region. G G V out V V Fig. 1 Basic common source amlifier biasing. The oerating oints of V and V out are the C or quiescent values at the inut and outut, resectively. deally, for a certain value of V, V out should always result in a corresonding value that does not change. However this relationshi will change due to temerature variations or manufacturing variability. Unfortunately the MOFET's ansconductance arameter cannot be conolled well during manufacturing. n addition, also varies with temerature. ome ansistor s can vary by more than a factor of 3. Note that is not directly secified in the data sheet but rather the ansconductance g m is secified. The relationshi between the MOFET s ansconductance g m and will be addressed in future labs. ince varies significantly, a circuit biased correctly for one ansistor may not be biased correctly for another ansistor, even for one with the same manufacture and art number. Therefore, a more robust biasing scheme than the one shown in Fig. 1 is needed, such that the MOFET's quiescent oerating oint is less sensitive to changes in. nsensitivity of the MOFET's quiescent oerating oint can be achieved by adding a resistor into the source branch of the circuit as shown in the Fig.. An analysis of this new circuit, similar to before, results in the following equations: V = V (3 G = ( V V = ( V V G = (4 ( V V ( V V Note that Eq. (4 can be maniulated into a quadratic for drain current = 0 (5

3 G G V V V out Fig. Basic common source amlifier with reduced sensitivity. Notice in Eq. (5 that when goes to zero (multily through by go to zero, Eq. (5 simlifies to: and take limit as = (6 G which is the same relationshi for the circuit in Fig. 1. (Why should this be exected? One the other hand, if the is large (relative to, then Eq. (5 can be aroximated by: (7 = 0 Notice that in Eq. (5, only aeared in one term, whose effect was minimized (relative to the other terms by a large value. The quadratic in Eq. (7 is a erfect square and can be rewritten as: = 0 or = (8 Thus if is large, is ractically indeendent of as desired. The general solution for valid for any is: = ( V V ( V V ( V V 1 1 (9 From Eq. (9, the aroximation of Eq. (8 can also be obtained by letting s get large (limit as s aroaches infinity.

4 The circuit of Fig. is still not most efficient because it requires two C ower sulies: one for V and another for V. The circuit in Fig. 3 remedies this. The Thévenin equivalent for the circuit consisting of V, 1, and in Fig. 3 gives the biasing circuit in Fig. where V = V th and G = th. With these Thévenin equivalents substituted into the circuit, the circuit is identical to the circuit in Fig. with the excetion that the bias voltage V is now deendent on V. The V voltage is now conolled by the roer choice of 1 and. This eliminates the need for a searate ower suly to conol V. The gate current into the MOFET is zero so the gate voltage is thus determined by only V and the 1 and voltage divider. 1 G V V out Fig. 3 Basic common source amlifier biasing with reduced sensitivity and emloying a single C voltage. The C Oerating oint of this circuit is stable for two rimary reasons: The gate voltage is determined only by the voltage divider 1 and, since insignificant current flows into the ansistor gate, and is therefore indeendent of the ansistor arameters (esecially. The source resistor stabilizes the C oerating oint through negative feedback. f increases for any reason, such as temerature change, the subsequent rise in source current increases the voltage dro across, thereby increasing V and thus decreasing V G (since the gate voltage V is constant. The decrease in V G will counteract the attemted increase in (.. PreLaboratory Exercise The MOFET you will be using in the lab is the ZVN3306A with a (in PCE secified at A / V (it can be as small as a third of this value and the nominal value of the threshold voltage is V = 1.8. But you have already estimated these values in Lab. Therefore use the values you estimated for comuting the circuit arameters in

5 the relab exercises. You are to use = 1000 Ω, = 470 Ω, and V = 1 V. Note that you must use Matlab to solve some of the relab roblems. etting Transistor Quiescent Point 1. erive the equation for the C load line for the MOFET circuit in Fig. 3 (with 1, and the source resistor used for added circuit stability by relating the current to the variables V, V,, and (irchhoff s voltage law. Be sure to sketch the circuit and the vs. V grah, which should contain sketched curves and a sketch of the load line, with end oints labeled symbolically (no numbers are required on the sketch or in the equation derived above. For an examle of these kind of curves see Figure 6.9 on age 33 in Horenstein.. etermine the quiescent oerating oint Q and V Q such that it is at the midoint of the C load line (to enhance the likelihood of a symmeical outut voltage swing. etermine voltage dros over and at this design oint. 3. Write a rogram in Matlab to lot the C load line suerimosed with the characteristic curves ( versus V of the Nchannel MOFET for an arbiary value of V G. Use mfiles from lab lecture to hel with this art. Modify your rogram to determine a value of V G iteratively, so the intersection of the characteristic curve and load line is close to the desired oerating oint. Hand in a rintout of the rogram (with comments, a lot of the load line and characteristic curve at the determined V G value, and indicate the value of V G (you can include this value in the title of the grah or simly handwrite it on the lot with clear indication of what it is. 4. Plot the drain characteristic curves measured in Lab with the load line. oes this grah suggest a similar value of V G as in the revious roblem? (You will have to use a rough estimation based on an interolation between the measured aces to determine what V G results in a ace close to your desire quiescent oint. iscuss the similarities or differences in the number you obtained. ecide which is more accurate. 5. For the quiescent drainsource voltages and current comuted in the revious roblems, use Eq. (6 to comute the gate to source voltage (V G at the designed quiescent oerating oint, and use Eq. (8 to comute the gatetoground voltage (V (same as voltage across. Comare and comment on this value of V G relative to the value you found by iteration in revious roblems. 6. etermine the ratio between 1 and that will result in the V value of the last roblem and maintain this oerating oint. n addition, determine actual values for 1 and that will also result in 1 = Q /100. Effects of Variations 7. Use the quiescent V comuted in the revious roblem to comute the resulting oerating oint ( Q and V Q for the cases when is 1/ and 1/3 its design value (using Eq. (9. PCE imulation 8. Create a PCE model for the circuit in Fig. 3 with the values given above and your comuted 1 and. Use the MOFET ZVN3306A model and change the and V values to values estimated in Lab. Perform a PCE oerating oint analysis to

6 determine the voltage from the drain to the ground and the drain current. Then use PCE oerating oint analysis to determine the drain source voltage (V Q and the drain current ( Q when has been decreased to 1/ and 1/3 of its design value. V. Laboratory Exercise 1. Measure quiescent oints: Consuct the MOFET common source amlifier (Fig. 3 using the Prelab values for 1,,, and V. Measure the circuit's quiescent oerating oint by measuring V Q, the exact value of and the voltage across or in order to calculate Q. Measure the quiescent oerating oint using the multimeter on the C settings. (iscussion: Comare these values with those redicted in the relab exercises. Observe effect of increasing ower suly voltage: ncrease V to 15 V and measure the circuit s quiescent oerating oint as in the revious rocedure. (iscussion: How did the quiescent oerating oint change and how did the circuit s load line change? s this what you would have exected? id the circuit's quiescent oerating oint stay at the midoint of the load line? Exlain.