II–3 DC Circuits II

  • Published on

  • View

  • Download

Embed Size (px)


II3 DC Circuits II. Applications. Main Topics. Example of Loop Currents Method Real Power Sources. Building DC Voltmeters and Ammeters. Using DC Voltmeters and Ammeters. Wheatstone Bridge. Charging Accumulators. The Thermocouple. Example IV-3. - PowerPoint PPT Presentation


  • II3 DC Circuits IIApplications

  • Main TopicsExample of Loop Currents Method Real Power Sources.Building DC Voltmeters and Ammeters.Using DC Voltmeters and Ammeters.Wheatstone Bridge.Charging Accumulators.The Thermocouple.

  • Example IV-3Let I be the current in the DBAD, I in the DCBC and I in the CBAC loops. Then:I1 = I - I I2 = I - II3 = I - II4 = -II5 = I I6 = I

  • Example IV-4The loop equation in DBAD would be:-V1 + R1(I - I) V3 + R3(I - I) + R5I = 0(R1 + R3 + R5)I - R1I - R3I = V1 + V3 Similarly from the loops DCBD and CABC:-R1I + (R1 + R2 + R4)I - R2I = V4 - V1 V2 -R3I - R2I +(R2 + R3 + R6)I = V2 - V3 It is some work but we have a system of only three equations which we can solve by hand!

  • Example IV-5Numerically we get: 12 2 5 I = 51 -2 14 10 I = -16 -5 10 25 I = 25From here we get I, I, I and then using them finally the branch currents I1

  • Real Power Sources IPower sources have some forces of non-electric character which compensate for discharging when current is delivered.Real sources are not able to compensate totally. Their terminal voltage is a decreasing function of current.Most power source behave linearly. It means we can describe their properties by two parameters, according to a model which describes them.

  • Real Power Sources IIMost common model is to substitute a real source by serial combination of an ideal power source of some voltage or EMF (electro-motoric force) and an ideal, so called, internal resistor. Then the terminal voltage can be expressed:V(I) = - RiIIf we compare this formula with behavior of a real source, we see that is the terminal voltage for zero current and Ri is the slope of the function.

  • Real Power Sources III can be obtained only by extrapolation to zero current. From the equation we see that the internal resistance Ri can be considered as a measure, how close is the particular power source to an ideal one. The smaller value of Ri the closer is the plot of the function to a constant function, which would be the behavior of an ideal power source whose terminal voltage doesnt depend on current.

  • Real Power Sources IVThe model using and Ri can be used both when charging or discharging the power source. The polarity of the potential drop on the internal resistor depends on the direction of current.Example: When charging a battery by a charger at Vc = 13.2 V the Ic = 10 A was reached. When discharging the same battery the terminal voltage Vd = 9.6 V and current Id = 20 A. Find the and Ri.

  • Real Power Sources VCharging: + Ic Ri = Vc Discharging: - Id Ri = Vd Here: + 10 Ri = 13.2 - 20 Ri = 9.6 = 12 V and Ri. = 0.12

  • DC Voltmeters and Ammeters IMeasurements of voltages and currents are very important not only in physics and electronics but in whole science and technology since most of scientific and technological quantities (such as temperature, pressure ) are usually converted to electrical values.Electric properties can be easily transported and measured.

  • DC Voltmeters and Ammeters IIIn the following part we shall first deal with the principles of building simple measuring devices. Then we shall illustrate some typical problems which stem from non-ideality of these instruments which influences the accuracy of the measured values.

  • Building V-meters and A-meters IThe heart of voltmeters or ammeters is so called galvanometer. It is a very sensitive voltmeter or ammeter. It is usually characterized by full-scale current or f-s voltage and internal resistance.Let us have a galvanometer of the full-scale current of If = 50 A and internal resistance Rg= 30 . Ohms law Vf = If Rg = 1.5 mV

  • Building V-meters and A-meters IIIf we want to measure larger currents, we have to use a shunt resistor which would bypasses the galvanometer and takes around the superfluous current.For instance let I0 = 10 mA. Since it is a parallel connection, at Vf = 1.5 mV, there must be I = 9.950 mA passing through it, so R = 0.1508 . Shunt resistors have small resistance, they are precise and robust.

  • Building V-meters and A-meters IIIIf we what to measure larger voltages we have to use a resistor in series with the galvanometer. On which there would be the superfluous voltage. Lets for instance measure V0 = 10 V. Then at If = 50 A there must be V = 9.9985 V on the resistor. So Rv= 199970 . These serial resistors must be large and precise.

  • Using V-meters and A-meters IDue to their non-ideal internal resistance voltmeters and ammeters can influence their or other instruments reading by a systematic error!What is ideal?Voltmeters are connected in parallel. They should have infinite resistance not to bypass the circuit.Ammeter are connected in serial. They should have zero resistance so there is no voltage on them.

  • Using V-meters and A-meters IILet us measure a resistance by a direct measurement. We can use two circuits.In the first one the voltage is measured accurately but the internal resistance of voltmeter (if infinity) makes the reading of current larger. The measured resistance is underestimated.Can be accepted for very small resistances.

  • Using V-meters and A-meters IIIIn the second scheme the current is measured accurately but the internal resistance of the ammeter (if not zero) makes the reading of voltage larger. The measured resistance is overestimated.Can be accepted for very large resistances.The internal resistances of the meters can be obtained by calibration.

  • Using V-meters and A-meters IVNormal measurements use some physical methods to get information about unknown properties of samples.Calibration is a special measurement done on known (standard) sample to obtain information on the method used.

  • Wheatstone Bridge IOne of the most accurate methods to measure resistance is using the Wheatstone Bridge. It is a square circuit of resistors. One of them is unknown. The three other must be known and one of the three must be variable. There is a galvanometer in one diagonal and a power source in the other.

  • Wheatstone Bridge IIDuring the measurement we change the value of the variable resistor till we balance the bridge, which means there is no current in the diagonal with the galvanometer. It is only possible if the potentials in the points a and b are the same:I1R1 = I3R3 and I1R2 = I3R4 divide them R2/R1 = R4/R3 e.g. R4 = R2R3/R1

  • HomeworkPlease, try to prepare as much as you can for the midterm exam!

  • Things to readRepeat the chapters 21 - 26 !

  • The vector or cross product I Let c=a.bDefinition (components)The magnitude |c|Is the surface of a parallelepiped made by a,b.

  • The vector or cross product IIThe vector c is perpendicular to the plane made by the vectors a and b and they have to form a right-turning system.ijk = {1 (even permutation), -1 (odd), 0 (eq.)}^