Electronic Circuits Laboratory EE462G .

Uploaded on:
Electronic Circuits Laboratory EE462G. Lab Safety, Experimental Measurements, and Statistics. Lab Safety. How much electricity does it take to kill or severely injure? Ø 0.01 A = Mild sensation, threshold of perception
Slide 1

Electronic Circuits Laboratory EE462G Lab Safety, Experimental Measurements, and Statistics

Slide 2

Lab Safety How much power does it take to execute or extremely harm? Ø 0.01 A = Mild sensation, edge of recognition Ø 0.1 A = Sever stun, extraordinary breathing troubles, can\'t give up, difficult Ø 1.0 A = Sever blazes, respiratory loss of motion, ventricular fibrillations

Slide 3

Lab Safety Turn off power supply and capacity generator when fabricating or adjusting a circuit. In the event that power supply is acting oddly or making parts extremely hot, don\'t utilize. Disengage control and convey it to the consideration of the TA or lab expert. Try not to stick confront close circuit while power is on. Eye insurance is suggested. Know about ground circles! Especially those made with your body. Try not to work with wet or damp hands. No nourishment or drink is allowed in the lab.

Slide 4

Lab Safety Leave the lab table as slick or superior to anything the way you discovered it. Clean up table top and mastermind hardware perfectly. Kill all gear (log off PCs) before clearing out. Move gear utilizing the handle or frame. Try not to move hardware by pulling on harmonies or tests. Try not to open up or dismantle gear. In the event that gear is missing or not working, convey it to the consideration of the TA or lab expert.

Slide 5

Lab and Instrumentation Terms Use these words in your review and exchanges! Precision Conformity of estimation to genuine esteem. Mistake Difference amongst measured and genuine esteem, regularly standardize by the genuine or expected esteem and given in percent. Accuracy Number of huge digits in estimation, relies on upon both the instrument determination and the repeatability of the procedure being measured. Determination Smallest quantifiable augmentation (reporting a number past the exactness or determination of the instrument is insignificant and misdirecting; the numbers won\'t be repeatable).

Slide 6

Lab and Instrumentation Terms Mean Error The normal blunder between a progression of measured qualities and the comparing genuine esteem (exactness/predisposition): is the i th estimation out of N autonomous estimations. is the valid or correlation esteem . is the inclination connected with the estimations.

Slide 7

Lab and Instrumentation Terms Average (Standard) Deviation The normal deviation of qualities from their mean shows the changeability of the estimation (compelling exactness): is the i th estimation out of N free estimations. is the example mean of the estimations is the specimen standard deviation.

Slide 8

Lab and Instrumentation Terms Percent Absolute Error Normalized mistake: If the genuine qualities are not known, you can\'t register blunder . Rather you register a distinction between 2 values that have comparative significance. For instance, you can figure the contrasts between the deliberate values in an examination and the esteem anticipated by hypothesis. At that point treat the esteem with the minimum expected blunder as the genuine esteem to think about hypothesis and examination.

Slide 9

Experimental Error Conclusions drawn from trial estimations must be tempered by the mistake/changeability/instability in the deliberate qualities. "Exploratory blunder" exists in estimations making instability because of varieties from the instrument determination and estimation repeatability. The test of good test plan is to utmost wellsprings of inconstancy in the estimations to minimize test blunder with respect to the question the trial is attempting to reply. Once the test is planned and estimations made, the exploratory mistake must be described keeping in mind the end goal to decide the level of instability in the information.

Slide 10

Measurements and Statistics Experimental estimations were performed where the normal or anticipated esteem was 10.2 Volts. Estimations were made under 3 unique conditions. The accompanying numbers were measured and recorded: Condition 1: 9.43, 10.5, 11.4, 10.33 Condition 2: 10.1, 10.2, 10.15, 10.14 Condition 3: 6.00, 5.72, 8.21, 6.33, 6.57 What estimation accuracy is recommended from the recorded numbers ? For which condition(s) are the distinctions in the estimation and the anticipated values probably because of test blunder.

Slide 11

Confidence Intervals Confidence Interval Estimation A certainty interim is an irregular interim whose end focuses  and  are elements of the watched arbitrary factors with the end goal that the likelihood of the disparity: is fulfilled to some foreordained likelihood ( ): Typically  = .05 for a 95% certainty interim. A helpful dispersion for mean qualities evaluated from a limited number of tests is the Student\'s t conveyance.

Slide 12

Student\'s t - Statistic Student\'s t Distribution Let be ordinarily dispersed genuine mean equivalent  genuine standard deviation rise to  (or difference  2 ) test mean example fluctuation Then the t - measurement gets to be with  = N - 1 degrees of flexibility and has the understudy\'s t - conveyance

Slide 13

=5 0.4 0.35 an Area = 1 - 0.3 0.25 an Area =/2 an Area =/2 0.2 0.15 0.1 0.05 0 - 5 - 4 - 3 - 2 - 1 0 1 2 3 4 5 X Student\'s t - Statistic Let irregular factors and signify values where the zone between these focuses under the thickness capacity is: Note for  =.05, the territory between t qualities is 0.95.

Slide 14

Computing Confidence Intervals Recall the t-measurement was characterized as: Therefore, the genuine mean is in an area (interim) of the specimen mean with likelihood (certainty) 1- :

Slide 15

t - measurement Table

Slide 16

Experimental Error Example Which conditions propose the contrasts between the deliberate and the anticipated qualities were not critical (@ 95% Level).

Slide 17

Repeating Measurements and Averaging If an esteem is being measured/assessed from a circuit, it ought to be measured various times , where wellsprings of inconstancy change autonomously from estimation to estimation, while the property of intrigue does not. The mean can be processed from the rehashed estimations as a gauge of the genuine esteem, and the standard deviation (or a similar measurement identified with variety) can be figured as a vulnerability appraise.

Slide 18

Curve Fitting to Sequence of Data Points If a capacity (or waveform) is of intrigue, then a succession of focuses over the capacity\'s space must be measured. The arrangement of area focuses must be sufficiently thick to determine basic components of the capacity. The capacity qualities ought to be measured different times with the property of intrigue held consistent. The mean of the deliberate capacity focuses at every space point can serve as a gauge. The standard deviation (or tantamount measurement) can be figured as a gauge of the instability . Certainty breaking points can be figured for every capacity point and included as mistake bars on the plot.

Slide 19

Parametric Curve Fitting If a grouping of measured focuses is relied upon to fit a specific bend controlled by a couple of parameters, (for example, an exponential, sinusoid, polynomial … .), then the mistake between the bend and the deliberate focuses can be registered as an element of the bend\'s parameters. The parameter set yielding the base mistake can then be utilized as the "best-fit" parameter appraise for depicting the information. The mistake between a capacity and an arrangement of focuses is regularly depicted by the mean square blunder (MSE): is measured information comparing to space point x i is the capacity planned to fit the deliberate focuses with parameter

Slide 20

Curve Fit Example Assume a progression of voltage yields because of a stage input takes after and exponential reaction of the shape: An analysis was performed where the framework was driven with a stage and the yields were measured now and again .01, .02, .03, .04, and .05 seconds. For 5 autonomous trials the relating yields were: time t=.01 t=.02 t=.03 t=.04 t=.05 Trial 1 0.5955 0.8719 0.8937 1.0293 0.8277 Trial 2 0.7014 0.8800 0.9685 1.0666 0.9740 Trial 3 0.6228 0.7757 0.8338 0.9324 1.0129 Trial 4 0.5942 0.7713 1.0253 0.9318 0.9925 Trial 5 0.6321 0.8472 0.9780 0.9632 0.8768 Write a Matlab script to locate the best estimation of α for every trial. At that point locate the mean, standard deviation and 95% certainty limits for the α gauges over all the autonomous trials. Plot the parametric bends in view of the mean α and those relating to the 95% certainty limits. Plot the deliberate information focuses on similar chart and contrast them relative with the 95% certainty interims. Remark on as far as possible for α in relationship to the genuine esteem (which was 100 for this situation).

Slide 21

Curve Fit Example Sample Matlab script to take care of issue and after that some additional: % Create vector relating to time focuses for every estimation t = [.01, .02, .03, .04, .05]; % Type in measured information focuses where every section compares to the % time point and every line compares to a trial. expmst = [0.5955 0.8719 0.8937 1.0293 0.8277; ... 0.7014 0.8800 0.9685 1.0666 0.9740; ... 0.6228 0.7757 0.8338 0.9324 1.0129; ... 0.5942 0.7713 1.0253 0.9318 0.9925; ... 0.6321 0.8472 0.9780 0.9632 0.8768]; % Type in a scope of alpha qualities to attempt in the bend fit process a = [50:.02:130]; % Vector from 50 to 130 in ventures of .02

Slide 22

Curve Fit Example % Loop to gauge alpha for every trial for kt=1:5 % (kt is the file for every analysis/estimation set) % Loop to figure the MSE for every alpha for ka=1:length(a) % MSE amongst information and parametric capacity is trial alpha esteem mse(ka) = mean((expmst(kt,:) - (1-exp(- a(ka)*t))).^2); end % Find least MSE and relating alpha for every estimation set [tmpval, tmppos] = min(mse);

View more...