Josef Pošta - Roman Pavlíček

Tel.: +4202/24383266, Fax: +4202/20921361, E-mail: Posta@tf.czu.cz

The alternator is the most spread rotary energy source used for vehicles and mobile machines. The machine availability depends on its technical state. Alternator’s failures have mechanical or electrical character. The mechanical failures of alternator can be detected and identified according to its operating during a work usually. The electrical failures are the best detected with a support of diagnostic facilities; according to external operations. In case of a collaboration of the alternator and accumulator (battery) under constant conditions the mild deviations of input parameters can be veiled and it is not possible to identify them reliably. The working assumption was that the neither small deviations is possible reliably identify by trend analyses of transiting process, occurring during the working change of alternator. From the point of view of equipment facilities the measurement with the oscilloscope card mounted into usual personal computer.

The measuring method of instantaneous voltage value on alternator’ clamps has been used; the alternator has been operating under the constant revolutions and load. If the alternator and its voltage regulator is perfect under such conditions then the d-c voltage obtained from a two-way directed voltage per every coil winding. Instantaneous values of the voltage are stated by constructional setting and alternator’s revolutions; expected trend is shown on fig. 2. In this case it is possible to consider the voltage changes in time interval as transiting processes. The time interval is comparable with expected duration of feeding output clamp change between two neighbouring alternator winding. The additional oscilloscopic cart PCX 1230 produced by TEDIA s.r.o. firm, was used for measuring. The card was installed in computer PC 386 SX, 25 MHz, 8 MB RAM. The controlling programme SCOPEWIN (delivered together with the oscilloscopic card) for used for operating and configuration of the card. The voltage divider was used as indicator. The indicator has been decreasing the output alternator voltage used as an input into oscilloscopic card. Resulting values were multiplied by dividing ratio of divider so the real numeric and graphical results respond the real values of voltage on alternator’ clamps.

Analogue signal was redirected as an input into A/D converter. The signal was converted into digital values saved into RAM memory of the card. Later on the values were transferred into PC’s memory. The sample A/D converter frequency is adjusted by software. The record duration (record length) is limited by RAM card capacity and by sample frequency.

Measured data was converted in a file with format *.xls and processed by MS EXCEL spreadsheet.

The large quantity of results was obtained under different operational conditions and different failures. The results of measurement are shown in this article for an illustration. The arranged failure was based on short of one diode from anode group. Expected trend (course) of voltage are shown on fig. 2. and fig. 4. The real voltage trend obtained by a referred way is shown on fig. 3. and 5.

According to published courses (and not published too) it is apparent that the real courses respond very well to expectations. Nevertheless the apparent effect of voltage swinging with six times higher period then the main harmonic component can be observed in results very well. It is apparent on tangential courses of maximal and minimal instantaneous values. The origin and technical causality of the phenomena has not possibly explain up to now.

The real values courses consist of points where the instantaneous voltage values were indicated. During the comparison of courses assumption (theoretical) with the real ones following conclusion can be accepted; the choice of sample frequency should be done to fulfil requirement for measuring of three values per time interval and voltage excess of one winding. On the base of the requirement it is possible to derive the formulas for minimal sample frequency:

(1)

where: f_m minimal sample frequency [Hz]

n revolutions alternator’s armature[rev/min]

v number of winding phases [1]

p computation magnetic pole of armature [1]

Maximum frequency of sampling is not restricted from the alternator’s point of view; it is limited by possibilities of used card, eventually PC.

During the described experiments additional knowledge was obtained. This knowledge is not fully proved and processed but it can serve as practical guide for diagnostic processes made by this way. All described items is possible to complete as follows:

• if the sufficiently high sample frequency is disposable; the deviation from a normal stage does not depend on revolutions of alternator’s armature so sensitively;
• current load of alternator has an influence on the method’s sensitivity. The sensitivity is decreased by the growing current;
• it is possible to easy diagnose the adjusting failures in voltage regulator. Semiconductor regulators have less differences of voltage excess then the vibration regulators.

The obtained results prove the good coincidence of experimental data with the assumptions. Since the assumptions are theoretically an experimentally found out and proved in such way that they are accepted as proper; the working hypotheses is evidently proved by the coincidence: alternator’s technical stage deviations is possible to safely recognise according to course changes, eventually position, instantaneous values of input voltage.

The carried out experiments proved that the indicating of instantaneous value is possibly sufficiently precisely and sensitively by a support of oscilloscopic card in personal computer. This settings have a range of advantages comparing the oscilloscope:

• if do not take into account the PC’s price, it is evidently cheaper;
• it enables comfortable data processing and data saving; the data can be used in a time changes observation of object;
• it is general-purpose and easily adaptable under very different requirements;
• it enables the visual interpretation of data

It is possible to conclude the article by the statement. The described method is suitable for completing diagnose of electrical failures of alternators under workshop (laboratories) conditions. The method (visual, easily repeating, savings data for non restricted time) is suitable for instructional and schooling purposes as well.