Lawrence V. Hmurcik

SPICE applications to an undergraduate electronics program

A plan is presented for integrating SPICE (simulation program with integrated circuit emphasis) into an undergraduate electronics curriculum. Three design projects for students are described: a network theory project, an analog electronics design course, and a senior thesis project. In the first project used in a first course on network theory, the emphasis is on simple programs that can extend basic techniques (such as Thevenins equivalent circuit) to cases where hand analysis is long and tedious. In the electronics course, the application of the bipolar-junction transistor and field-effect transistor is more clearly understood by the student when he/she constructs an analog amplifier according to a given set of specifications. Creativity is encouraged, since the design of the amplifier is flexible as long as the specifications are met. In the senior thesis project, SPICE, along with other computer design packages, enables the student to analyze and modify a project before it is constructed. This saves valuable construction time and allows for a more thorough theoretical analysis. >

Solitons created by chirped initial profiles in coherent pulse propagation

If an incident pulse is chirped, the critical parameters for self-induced transparency to occur in coherent pulse propagation can no longer be obtained from the well-known McCall-Hahn area theorem. We have been able to obtain these parameters by solving the Zakharov-Shabat eigenvalue equation for the bound-state eigenvalues. We find that critical (threshold) areas will be increased for a chirped incident pulse in almost all cases, except for a box profile or for a pulse that is approximately box-like in shape. In these latter cases, the chirped critical areas will instead decrease for the second and all higher branches. The first branch’s critical area is always increased due to chirping.

The effects of heat treatments on the transport properties of CuxS thin films

We have studied the effects of heat treatments on three CuxS thin films (1.995≤x≤2). Our results suggest that initial heat treatments cause copper in grain boundaries to diffuse irreversibly into the CuxS crystallites. Subsequent heating in hydrogen causes a reduction in surface oxides while the reverse process occurs in an oxygen atmosphere. At a given elevated temperature, the resistivity ρ and charge density P vary with time according to the expressions P=P0e±(t/τ)1/2 and ρ=ρ0e∓(t/τ)1/2 . On the other hand, the mobility is found to be approximately constant at a given temperature during heat treatment.

Frequency dependent capacitance studies of the CdS/Cu2S thin‐film solar cell

Measurements of the dark capacitance of CdS cells as a function of the frequency of the applied signal voltage have shown that the capacitance varies with frequency according to the relation C−C∞ = (C0−C∞)/(1+ω2τ2), where τ is the time constant associated with interfacial and deep bulk states. Photocapacitance studies show that the CdS cell can be treated according to the frequency dependent model of Schibli and Milnes. Under the influence of light, C∼1/(ω)1/2. Further work demonstrates that the simple planar junction model most often used to describe the CdS cell is accurate at high frequencies.

Deep trap levels in CdS solar cells observed by capacitance measurements

This paper is concerned with ac capacitance measurements taken as a function of reverse bias voltage and signal frequency on four thin‐film CdS/CuxS solar cells and one single‐crystal cell. Our results suggest the presence of at least two trapping states in the i layer formed by the presence of Cu impurities in CdS. The deep traps produce anomalies in the normally linear 1/C2 vs V plots. The energies of the states were found by measuring the particular bias voltages at which anomalies were observed. Other results concerning the anomalous behavior in CdS cells and their relation to the work of Roberts and Crowell are also discussed. These include variations of capacitance with frequency and temperature, and changes in shunt conductance with frequency.

A damage mechanism: lightning-initiated fault-current arcs to communication cables buried beneath overhead electric power lines

A lightning strike to an overhead structure will cause a brief arc through the soil from its lightning ground to any nearby grounded metal utility line such as a gas pipe, sewer line, or communications cable. A limited amount of damage to the buried line may result from such a stroke. However, if the overhead structure happens to be an energized conductor of an electric power line, the situation becomes dangerous: the lightning impulse will establish a conductive path across the power line insulator, down the pole and through the soil to the buried utility line. In a significant number of cases, this conductive path will allow the establishment of a large, long-duration power fault current from the lightning struck power conductor to the buried utility line. This power arc will terminate on the grounded pipe or cable shield, causing rupture and failure. The existence of this damage mechanism was confirmed in the laboratory with a full scale mock-up of a utility right-of-way. The phenomenon of lightning-triggered arc establishment through soil was then examined more closely with a high resolution apparatus in which most parameters could be tightly controlled. Artificial lightning impulses from 0.3 MV to 2.8 MV and 60 Hz power line voltages from 6.24 kV to 15.71 kV were used. Soil condition, electrode spacing, power line voltage, lightning impulse voltage and geometry were found to govern the probability of a lightning-initiated fault current arc through the soil in a predictable manner. For soil of 500000 /spl Omega/-cm resistivity, the distance between a simulated power system lightning ground and a buried cable at which a fault current arc is not initiated was found to be about 40 cm. This safe distance was proportional to the geometric mean of the power line voltage and the peak lightning impulse voltage.

Linear regression analysis in a first physics lab

Linear regression analysis (least squares) is used in the first physics lab in order to introduce students to computer‐aided analysis and to teach data fitting techniques. Application is made to two experiments: Fletcher’s trolley and Hooke’s law. Least squares will extract information from raw data in a very precise way, and it opens the way for the study of more complicated phenomena than a first lab usually covers. Students who learn least squares in the first semester do much better in future labs, including nonphysics labs.

Van der Pauw measurement of metal fibre orientation in a plastic-metal composite

Stainless steel fibres in ABS plastic form a composite with an anisotropic resistivity. Samples are rectangular shapes with uniform thickness. By assuming two principal resistivities and by using van der Pauws technique, we findϱ2 =ϱxϱy. For rectangular samples, field theory determinesϱy/ϱx and hence fibre direction. Results for three sample geometries agree with the theoretical predictions of the fibre patterns and with X-ray data. Samples formed by a centre-sprue feed are the best for fabricating large, uniform samples, while samples with a large length-to-width ratio have the most uniform metal density and fibre orientation. Resistivity was also measured by the more-common two-probe technique. Results correlate well to van der Pauw data, with 95% confidence.

Van Der Pauw Measurements of Resistivity in Plastic/Metal Composites

The van der Pauw (VdP) technique provides a four-probe measurement of a samples average resistivity. It is generally applied to samples which have a uniform density and thickness but an arbitrary shape. We apply the VdP method to a metal/plastic composite. Samples are cut into rectangular shapes, but internally the metal structure is a random shape. Values of resistivity correlate well (> 90%) to values measured by a two-probe technique. We can determine a composites structure in a highly accurate though destruc tive manner. Some in-situ test applications are proposed.

Using the Fourier series in a circuits or DSP laboratory

Laboratory experiments on circuits that exhibit transient effects are usually performed with a periodic input signal (like a square wave). Three experiments on a resistor capacitor (RC) filter using transient analysis and the solution of a first-order differential equation are described. The results according to a Fourier analysis are then reinterpreted. The similarities and differences in the results from the two theories provide a better understanding of both. Also, using Fourier analysis introduces the basic strengths and weaknesses of discrete signal processing. Further additions to this laboratory are discussed and include (1) the addition of more first-order linear circuits; (2) the addition of second-order (and higher-order) linear circuits; (3) the processing of all data in the frequency domain; and (4) the study of nonlinear circuits.