WO1993014411A1

WO1993014411A1 - Method and apparatus for non-destructive testing electric insulation by measuring its voltage response signal

Info

Publication number: WO1993014411A1
Application number: PCT/HU1993/000002
Authority: WO
Inventors: Endre NÉMETH
Original assignee: Budapesti Mu^'szaki Egyetem
Priority date: 1992-01-21
Filing date: 1993-01-21
Publication date: 1993-07-22
Also published as: HU207897B; HU9200183D0; EP0623218A1

Abstract

The invention relates to a method for non-destructive testing electric insulations by measuring the direct voltage response by connecting to the insulation direct voltage and measuring the decay voltage and the recovery voltage. In the sense of the invention in the first step the measuring direct voltage (Uo) will be connected on the terminals of the insulation to be tested for a charging time period (tt) depending on the time constant of the polarization processes in the said insulation, after expiration of the charging time period (tt) in the second step the measuring direct voltage (Uo) will be switched off and the initial slope (Mk) of the decay voltage (Uk) will be determined, in the third step said insulation will be short-circuited for a discharge time period (trz) depending on the time constant of the polarization range to be investigated, following the discharge time period (trz) the short circuit will be interrupted and the initial slope (Mv) of the recovery voltage (Uv) will be determined. The invention relates to furthermore an apparatus for non-destructive testing electric insulations by charging the insulation to be tested to direct voltage and subsequent discharging said insulation, wherein to the insulation to be tested the voltage source producing the measuring direct voltage is connected through the series connected switch element. According to the invention between the insulation to be tested, i.e. the sample tested (Pr) and the direct voltage source (G) a first controlled switch element (K1) is series connected, to its outlet connected to the sample tested (Pr) one outlet of the second controlled switch element (K2) coupled parallel with the sample tested (Pr) is connected, while parallel with the sample tested (Pr) a non-consumption voltmeter (F) and/or recorder is connected.

Description

Method and apparatus for non-destructive testing electric insulation by measuring its voltage response signal.

BACKGROUND OF THE INVENTION

Field of the invention

The invention relates to method and apparatus for no destructive testing electric insulation by measuring t total voltage response signal. The invention should be us advantageously for testing high voltage insulations well.

The testing and measuring system according to the inventi has been based on measuring the total voltage respon signal, i.e. on measuring the decay voltage diagram and t recovery voltage diagram as the function of the time and measuring the initial slope of said diagrams referring the unit charge voltage .These ^' data measured are used f characterizing and testing the condition of the insulations

The portable apparatus operating on the basis of the metho according to the invention can be used for diagnosti testing insulation system in the low voltage and high vol tage equipments on the site and without destroying sai system.

The task of an electric insulation is alwyas to separat electrically the metal elements of an equipment bein connected to different voltages. However the insulation ca fulfil reliably the task set only if the electric strengt of said insulation has been made safety higher than th electric stress occuring in operation.

The electric strength of the insulating materials used i the insulations can not be regarded to be constant. Th value of the electric strength is being continous decreasing from its initial value referring to the n insulation on the effect of the chemical and physic deteriorating processes, ageing etc., occuring during t operation.The decrease of the electric strength effects decrease of the electric safety of the operation of sa insulation system.

Progressing the deterioration processes to preset conditio the safety operation of said insulation and such that of th apparatus comprising said insulation becomes unsafe, whi results in sudden increase of the probability e.g. t frequency of punctures or breakdowns.

The puncture and breakdown during the operation of an ap paratus or machine causes unexpected operating troubles fo example longer or shorter failure or absence of the main supply voltage, which causes troubles and inconvenience no only to the civilien population, but the working machine will stop too, decreasing the production or casusing troubl for example in the traffic on the streets.

Such absence of the mains voltage could have seriou consequences for example in the metallurgical or chemica works, as any unexpected stop of the chemical processes cos always a lot of money too.

The real value of the electric strength of the insulation in operation could have been determined by measurements onl which destroy the insulation at the time of the measuring As such this way of measuring the electric strength can no be used in insulations being in operation.

However there is a demand for non-destructive testing th deterioration, descrease or change of the electric strengt during operation in order to avoid unexpected breakdowns an to decrease the danger of unexpected break of the suppl voltage.

The deterioration processes of the insulation in operati influence in most of the insulating materials not only i dielectric strength but also the conductivity and t polarization processes of long time constant (T > 0,1 sec. range.

Some decades ago people looked forward with gre expectations to determine the conditions and state of t insulation by measuring the loss factor at industri frequencies. However this method has not fulfilled t expectations, as by investigating the loss factor industrial frequencies only the polarization processes o much smaller time constant could have been investigated tha the time constant range mentioned above, and which has no been influenced by the deterioration of the insulatin material.

The change of the insulating resistance measured in th period, for example in one minute following switching on th measuring voltage can characterize in some cases, fo example in case of impregnated paper insulation, the dete rioration process. The disadvantage of this measuring metho lies in that the value of the insulation resistance measure e. g. after one minute will be influenced not only by th condition of the insulation that on one hand by th polarization processes in the material on the other hand b the geometrical form of the insulation as well, it mean that the insulation resistance measured can be ver different in dependence on the form and dimension of th insulation to be tested.

Experiments have been made to find an factor independen from the dimensions of the insulation by measuring th current flowing through the insulation and to characteriz the progress of the deterioration. Such factor is e.g. th so called absorption factor

^{or tl e} ratio C/R i.e e /^ . Although these factors are independent rom the dimensions of the insulation, disadvantegous is that the value of these factors have been influenced both by conduction and polarization processes in the insulating material.

It could happen that the two processes, namely the con¬ duction and the polarization, have opposite affects and such diminishing or even ceasing the affect of each other, so on the basis of this measurement one cannot reliably determine the progress of the deterioration of the insulation.

Prior art

This measuring method has been described by Gy.Vajda: "Deterioration of Insulations and their Investigation." Nowadays the conduction and polarization are examined by applying direct voltage to the insulation to be measured and either the charging leakage current or following the switching off the voltage the so called discharge leakage current will be measured in the function of the time. These are essentially the current response signals of the insulation given on the change of the electric field intensivity. Although by means of this method the conduction and polarization could be examined, however the current to be measured are very small (10^~8...10^~12 A) and as such needs highly sensitive measuring instruments, which are sensitive on the electrical noises effected by the electric field of the mains, as well, so using this method under industrial conditions seems to be very difficult.

Of course the current to be measured by means of the method mentioned above should be increased by increasing the measuring voltage, the value of the measuring voltage is limited on one hand by the safety operation on the other hand by the electric strength of the apparatus investigated or by the dielectric strength of the air in the measuri circuit (i.e. starting partial discharges on the met conductors and electrodes) , and the requirement that t test voltage never should risk or endanger the insulation be measured. On the basis of the above the measuring volta in general never should be higher that 5 kV or the proof operating voltage of the insulation to be measured.

Testing the insulation on large machines another probl could arise, namely during the measuring process the samp is in electric connection through the measuring volta source with the mains too, and as such the supply volta of the measuring apparatus^' has to be filtered a stabilized, in order to avoid the leakage current to influenced by the noises coming from mains or by th capactive current originating from the mains voltag variation.

Owing to the difficulties mentioned above measuring th current response signal for examining the operatin insulation has not been used. Mr Vajda in his book cite earlier describes this fact too.

In the last decades you find some indication that th voltage response signal of the insulation on the change o the field intensivity, i.e. the time function of the deca voltage and the recovery voltage measured on the terminals o the insulation are suitable for testing of the progress o deterioration processes of insulating materials b investigating the conduction and polarization. This i described in the journal Elektrotechnika 1966 No 1-2 o pages 21-23. The same matter is described by Endre Nemeth

Testing the insulation without destruction by measuring th discharge and recovery voltage, issued in the publicatio on the IX internat. Wiss. Koll TH Ilmenau. At the separate measurement of the decay voltage Ujζ(t) and the recovery voltage U_v(t) long short circuit periods are necessary for the neutralization of the polarizaton processes between the two measurements. This measuring takes much time, and it can be applied only in laboratorial circumstances. As operating measuring, only the the recovery voltage was determined, mainly in the case of testing the transformer insulations. In connection with the above, Endre Nemeth describes an article titled "Non-Destructive Testing Methods for Insulations" in the journal named "Electrotechnics 79" in 1986, in the item 1-2, on pages 12- 14., and resp. J. Schmidt wrote an article titled "Remarks to the Statement of dr. E. Nemeth" also in the mentioned document Electrotechnics on pages 14-15. Furthermore, Alajos Bognar et. al. describes an article titled "Testing High- voltage Oil-Paper Insulation . System Using Direct Current Dielectrical Field " in the document CIGRέ in 1990, No. 15, on pages 33-38.

Summary of the invention

The aim of the invention lies in developing the method and apparatus, using also the measuring the voltage response, i.e. the measuring of the decay voltage and the recovery voltage, however, the object of the invention is to eliminate the mentioned disadvantages, i.e. to shorten significantly the measuring time period, and to enable its using in industrial circumstances as well.

It has been recognized that if the initial slope of the voltages is applied for characterizing the state of the insulation, it is enough to measure only the initial part of the slopes. This time, the whole voltage response of the insulations can be determined in a single measuring method if the requested conditions is ensured,i.e. during measuring the slope of the decay voltage U_j , the voltage on the electrodes of the insulation does not decrease under the 90% of the measuring direct voltage U₀, as charging voltage. T mentioned decrease slightly influences the just aft executed measuring value by means of the recovery voltag The extent of the influence depends on the charging ti period t-t, the discharge time period t_rz, the slope of t decay voltage, and the time constant of the polarizati range. Generally the influence is slight in operati measurements, in particular cases the extent of t influence can be determined and the effect can be improved

By this method the period of the measurement can considerably shortened by for example 80-85 %, therefor the method can be applied for checking operating insulatio as well. The previously used measuring method comprises t following steps: the charge had been taken place for t charging time period t*t, Uj- was the measuring slope of t decay voltage, then followed the discharge of the sample f a period of 5 x t , then a further charge took place f measuring the recovery voltage U_v. In practice it means th to apply the measurement takes 6-7 times more charging ti period. Considering the method according to the inventio only 1,2 times more charging time period is requested.

It has been recognized in developing the apparatus accordin to the invention that the influence of the measurin instrument to the value of the measuring can be disregarde only in the case, if its inner insulating resistance is tw scales higher than that of the sample tested. In the ligh of the above mentioned fact, naturally, in the measurin instrument a voltmeter of high resistance, (rotor voltmeter or electrometer with oscillatory condenser) , and specia switching elements of high resistance, Reed-relay of hig voltage or vacuum-switch are applied.

Description of the preferred embodiments Thus the invention relates to a method for non-destructiv testing electric insulations by measuring the direct voltag response by connecting to the insulation direct voltage an measuring the decay voltage and the recovery voltage.

The essence of the method according to the invention is i that in the first step the measuring direct voltage will b connected on the terminals of the insulation to be teste for a charging time period depending on the time constant o the polarization processes in the said insulation, afte expiration of the charging time period in the second ste the measuring direct voltage will be switched off and th initial slope of the decay voltage will be determined, i the third step said insulation will be short-circuited for discharge time period depending on the time constant of th polarization range to be investigated, following th discharge time period the short ciruit will be interrupte and the initial slope of the recovery voltage will b determined.

It is advantageous in the method according to the invention, if the discharge will be continued until the measured deca voltage decreases to the not more than 90% of the measurin direct voltage.

It is also advantageous, if the measuring direct voltage i lower than the operational voltage of the insulation, optinonally the measuring voltage range is 1 kV...5 kV.

The method according to the invention ensures the measuring conduction and polarization in the insulation, i.e the character of the condition of the insulation will be characterized by the initial slopes measured.

The method according to the invention can be applied so that the whole measuring process will be repeated regularly or irregularly, the measured values will be registrated, and the condition and state of the insulation will be watched a controlled on the basis of the data measured.

It is advantageous furthermore, if the measuring dire voltage is lower than the voltage being dangerous for t insulation.

The invention relates to furthermore an apparatus for no destructive testing electric insulations by charging t insulation to be tested to direct voltage and subseque discharging said insulation, wherein to the insulation to tested the voltage source producing the measuring direc voltage is connected through the serious connected switc element.

Thus the essence of the apparatus according to the inventio is in that between the insulation to be tested, i.e. th sample tested and the direct voltage source a firs controlled switch element is serious connected, to it outlet connected to the sample tested one outlet of th second controlled switch element coupled parallel with th sample tested is connected, while parallel with the sampl tested a non-consumption voltmeter and/or recorder i connected.

According to a preferred embodiment of the invention th control input of the first switching element is connected t the element producing the charging time period, while th control input of the second switching element is joined t the element producing the discharge time period, which ar arranged in the control unit realizing the requested, i itself known, timed operation of the first and secon switching elements.

Short description of the drawings The invention will be described in details hereunder by th aid of the drawings, illustrating some embodiments by way o example of the apparatus realized by the aid of the metho according to the invention. In the drawing:

Fig.l. shows the measuring principle of the measurement according to the invention representing the schematic circuit diagram.

Fig.2. represents the time function of the voltage to be realized on the course of the measuring method according to the invention,

Fig.3 shows the detailed embodiment of the apparatus according to the invention.

In Fig. 1 can be seen the direct voltage source G, one of its outlets is connected to one of the outlets of the first switching element Kl of high voltage, the other outlet of the first switching element Kl of high voltage connects to one of the outlets of the second switching element K2 of also high-voltage, while the other outlet of the second switching element K2 joins to the other outlet of the direct voltage source G. Parallel with the second switching element K2 the sample tested Pr symbolizing the insulation to be examined and a voltmeter - non-consumption voltmeter - are coupled.

In Fig. 2. can be seen the the time function to be realized by the method according to the schematic circuit diagram according to the Fig. l. According to Fig. 2. the measuring direct voltage U₀ by closing the switch Kl should be connected to the sample tested Pr for charging time period -fc then the measuring direct voltage U₀ by opening the switch Kl is disconnected and the voltage will be discharge on the sample tested Pr. This slope of the decay voltage U^, is the initial slope of the decay voltage, as the discharg is performed up to the 90% of the measuring direct voltag U₀, i.e. on the sample tested Pr only a voltage decrease o

10% can be taken place meanwhile determining the initia

^• slope Mj . After that the sample tested Pr by closing th switch K2 will be short circuited up to the discharge tim period t_r2, then the short circuit will be interrupted, an the recovery voltage U_v is measured and the initial slope M is determined.

In Fig. 3. shows in details the circuit diagram of th apparatus realizing the method^" according to the invention, wherein a supply source T can be seen producing the measuring direct voltage U₀ and which produces the stable direct voltage of a previous given value from the mains voltage by the way known in itself, and the signal comprises the transformer Tr separated by bipolarly, and a rectifier El series connected to the output of the transformer. The output measuring direct voltage U₀ of the supply source T is connected to the first switching element Kl and the second switching element K2 in the way shown in Fig. 1, wherein the first switching element Kl and the second switching element K2 are controlled switches of high-voltage with high resistance, the control coins of which is connected to the output of the control unit V producing the charging time period t--- as well as the discharge time period t_rz. The control unit V is not described in detail, while it can be realized in every case from the elements known in itself, as the charging time period t-t and the discharge time period t_r2; can be adjusted in the range given in advance necessary for the operation of the first switching element Kl and the second switching element K2 have to be ensured. The first switching element Kl and the second switching element K2 form together a switching unit K, the output of which is connected to the voltmeter F; this voltmeter contains the non-consumption voltmeter element Fl, and the A/D converte Al joining thereto, which connects to the control unit V. The output of the voltmeter F is connected to the insulatio to be tested.

In the figure can also be seen the display M, which i connected to the control unit V, which stores the data an ensures not only the display of the results but the handlin of the measuring instrument. ^'

The operation of the method resp. apparatus according to th invention is as follows:

In the first step, the measuring direct voltage U_Q has to be determined suitable for the insulation to be tested. The measuring direct voltage U₀ has to be determined so that it should not be harm to the insulation, but its value should be enough for developing the dielectic processes. According to our experiences the value of the measuring direct voltage U₀ depending on the operation voltage of the insulation to be tested is optinonally between 1 kV...5 kV. Should the value of the measuring direct voltage U₀ is available, in the first step, the first switching element Kl is closed up to a previously given charging time period t-^, then the charging time period t^ can be determined in the way known in itself, in the knowledge of the time constant of the polarization processes having in the insulation to be tested.

In the second step, after the charging time period t-*- elapsed the first switching element Kl is opened, so disconnecting it from the direct voltage source G producing the measuring direct voltage U₀, and by the aid of the voltmeter - which is non-consumption, for example electrostatical voltmeter - the slope Mjς of the decay voltage U^ on the electrodes of the sample tested Pr is measured resp. recorded. The measurement is taking place up to the time, when the initial slope M^ of the decay voltag Ujζ can be determined.

In the following step, the electrodes of the sample teste Pr resp. the insulation to be tested will be short circuite with the second switching element K2 for a discharge tim period t_r2 determined in advance, i.e. the charge accumulated on the electrodes of the insulation will b discharged. The period of the discharge time t_rz depends o the time constant of the polarization range to be tested however the data can be known from the respectiv bibliography.

After the discharge time period tt elapsed the secon switching element K2 is opened, so the short circuit of th sample tested Pr (resp. the insulation to be tested) i interrupted, and the slope of the recovery voltage U_v bein on the electrodes of the insulation is measured resp recorded. In this case the measurement is taking place als up to the time, when the initial slope M_v of the recover voltage U_v can be determined.

The initial slopes ^ resp. M_v determined from the point of the slope of the decay voltage ϋ _ and the slope of th recovery voltage U_v are expedient recorded.

The intensity of the conduction resp. polarization processe having on the insulation - the condition of the insulatio due to the deterioration processes - can be characterized b the slopes \H _ resp. M_v (relating to the unit chargin voltage) of the decay voltage U^ resp. of the recover voltage U_v. After the measurement the insulation has to b short circuited for safety reasons.

The above mentioned four steps, i.e. the charging, th measuring the slope of the decay voltage Ujς, the shor circuiting of the insulation, the measuring the slope of th recovery voltage U_v, provide the same effect, inspite of th essentially shorter measuring period requirement, with th prescribed conditions and measuring result applied until no by the method. The method applied upto this time determine the initial slope of the voltages in two measurement following each other, by the aid of short circuited of th insulation for longer period as well as by means of resting.

The method according to the invention concerning measurin the slopes of the decay voltage Uj resp. the recover voltage U_v has the following advantages in contradiction t the method applied until now:

The measuring direct voltage U₀ used for testing insulation is lower than the previously used one, i.e. its value range is 1...5 kV which on one hand fulfills the safety requirements, on the other hand-, no discharge is taken place by the applied measuring voltage, so, does not disturb the measurement not deteriorating the insulation.

The measurement of the slope of the decay voltage Uj resp. recovery voltage U_v, as voltage response signal, needs only the voltage of the value from some V to 100 V and similarly, the slope of the recovery voltage U_v is measured in the same range of order. Measuring these voltages causes smaller measuring technical problems, just are less sensitive for the outer electric disturbances on contrary to the previously used method, particularly in the case, if small currents were measured.

During measuring the voltage response of the insulation the sample tested Pr resp. the insulation have been disconnected from the supply source T or from the direct voltage source G, so the possible fluctuation of the mains voltage, the other disturbances coming from the mains voltage do not influence the result of the measurement against with the method measuring the current. The connection of the testing voltages preceding the test of the insulation as well as t voltage fluctuation forming during charges cause no effe on the result of the measurement.

Measuring the slope of the decay voltage Ujς and the slope the recovery voltage U_v make possible to test separately t conduction as well as the polarization processes, by th means, in some cases, for example using the impregnat paper insulation affords possibility for separate testing the progress of deterioration processes as well. It can realized by the fact that the initial slope of the dec voltage U^ resp. recovery voltage U_v referred to un charging voltage, are directly proportional with t intensity of the conduction resp. polarization process being in the insulation, resp. with the specif characteristics concerning the intensity of the processe Therefore, on one hand, by measuring the voltage response the followings can be determined: the specif characteristics (per unit volume) of the insulati referring to the polarization resp. to the conduction, the other hand, due to the mentioned direct proportion, t initial slopes of the voltages M_jς resp. the M_v can applied directly as a measuring number, characterizing the condition of the insulation, so by means of their using the results measured on the insulations of differen dimension and geometry, can be direct compared an appraised.

The method according to the invention can be applied fo testing electric insulation of high-voltage, such a insulations for rotors, transformers as well as cables.

In the apparatus according to the invention the controlle can be of the control unit V with both computer an microprocessor as well, however, it cannot be described now as it is known in itself. As far as the first switchin element Kl and the second one K2 are concerned, they ar switching means of high resis-tance resp. they have to b controlled suitably. They can be of various switches wit semi-conductor or for example vacuum- or Reed switsches.

Claims

1. Method for non-destructive testing electric insulations measuring the direct voltage response by connecting to t insulation direct voltage and measuring the decay volta and the recovery voltage, characterized by that in the fir step the measuring direct voltage (U₀) will be connected the terminals of the insulation to be tested for a chargi time period (t-t) depending on the time constant of t polarization processes in the said insulation, afte expiration of the charging time period (t-^) in the secon step the measuring direct voltage (U₀) will be switched of and the initial slope (M^) of the decay voltage (Uj) will b determined, in the third step said insulation will be short circuited for a discharge time period (t_rz) depending on th time constant of the polarization range to be investigated following the discharge time period (t_rz) the short circui will be interrupted and the initial slope (M_v) of th recovery voltage (U_v) will be determined.

2. Method according to Claim 1 characterized in that th discharge will be continued until the measured decay voltag (Uj) decreases to the abt. 90% of the measuring direc voltage (U₀) .

3. Method according to the Claims 1 or 2, characterized in tha the measuring direct voltage (U₀) is lower than th operational voltage of the insulation, optinonally th measuring voltage range is 1 kV...5 kV.

4. Method according to any of the Claims 1 to 3, characterize in that the conduction and polarization in the insulation, i.e the character of the condition of the insulation will b characterized by the initial slopes (Mjς,M_v) measured.

5. Method according to any of the Claims l to 4, characterize in that the whole measuring process will be repeate regularly or irregularly, the measured values will b registrated, and the condition and state of the insulatio will be watched and controlled on the basis of the dat measured.

6. Method according to any of the Claims 1 to 2 and 4 to 5, characterizedin that the measuring direct voltage (U₀) is lower than the voltage being dangerous for the insulation.

7.Apparatus for non-destructive testing electric insulations by charging the insulation to be tested to direct voltage and subsequent discharging said insulation, wherein to the insulation to be tested the voltage source producing the measuring direct voltage is connected through the serious connected switch element, characterized in that between the insulation to be tested, i.e. the sample tested (Pr) and the direct voltage source (G) a first controlled switch element (Kl) is seriously connected, to its outlet connected to the sample tested (Pr) one outlet of the second controlled switch element (K2) coupled parallel with the sample tested (Pr) is connected, while parallel with the sample tested (Pr) a non-consumption voltmeter (F) and/or recorder is connected.

8. Apparatus according to the Claim 7, characterized in that the control input of the first switching element (Kl) is connected to the element producing the charging time period t-t) , while the control input of the second switching element (K2) is joined to the element producing the discharge time period (t_rz) , which are arranged in the control unit (V) realizing the requested, in itself known, timed operation of the first and second switching elements (Kl, K2) .