ELECTRO?OPTICAL MEASURING APPARATUS

M The present invention relates to method and a device for operating the flow of signals in a photoelectron measuring device, wherein there is applied to the electronic evaluating means, via two parallel channels, electrical signals quadrature may be described essentially of the functions

ijL (t)* ^ /" 1 + sin (0 (t) W) _7 (1)

and

±₂ k (t)-₂ £1+ cos (0 (t) (2)

where i_{n 0} represents the current variable as a function of time t,

kl 2 are constants, 0 (t) is proportional to the measured quantity, and ' \ ^ is a constant of base_e

Photoelectron A measuring device wherein electrical signals according to the equations (1) and (2) are obtained by interferometric intensity modulation of a light beam, is known for example by the Swiss patent 065-433

For example, by the IEEE J, * El The_e QE-2, no. 8 (1966), page 255" it is known to influence the inclination of the plane of oscillation of a laser light beam, linearly polarized, photoelectron in a measuring device, in proportion to the measured quantity, using a magneto-optical crystal, and then splits the light beam, using a Glan- ïhomsen modified, into two partial beams, with the direction of oscillation are offset from one another by ∩Τ/2, thereby, after converting light intensities into electrical currents, d * obtain signals which are in principle the shape of the equations (1) and (2), but with a phase shift. According a request for, patent 70-45 999 Prance 21-12-70 deposited in the by the present applicant is aware, this known method is modified such that the directions of oscillation of the two partial light beams are déasalées each other ' If / 4, thereby attaining electric signals which correspond to the equations (1) and (2), also in relation to the phase shift.

While in this application of " said patent, is used an electronic device operating, is substantially similar to that of the known solutions mentioned in the preamble, in a second patent application no. 71,02105 Prance 22.1ο71 deposited in the by the present invention, designated another attractive process electronic operating.

the electrical signals according to the solutions above recited are described with a good approximation by the equations (1) and (2). Hereinafter, the constant factor to the sine or the cosine, which appears with the interferometric modulation, is substantially equal to a ( J.Opt. Soco Am. (1957) > 1097-1105), and quadrature errors or zero, which can be represented by a constant phase, variable as a function of time, may be disregarded.

With these solutions above mentioned, is a constant phase '-vp " 0. Therefore, the signal Ig, for a non-linear function (t) 0 of the time t, is different from the signal i-^, and, unlike the signal i ^ it has a direct current component.

The measurement accuracy measuring devices photoelectronic types indicated is strongly influenced by the factors-^ k g in the equations (1), (2), which determine the amplitudes of the signals, are or are not at all times strictly equal to each other * Furthermore, in the electronic evaluating means, to subtract signals the correct value¹ g,^the QtJfait disappear the "1" which is in parenthesis; it is indispensable, therefore, know at all times accurately the exact value of k^g"^Ges problems are solved, in the known solutions, and that is, as a consequence, the object of the invention to provide a method and a device, for maintain at all times the equality k ^ * kg, and the subtracted signals always i-^ g the airport value-

te k. ^. This then allows k variations.₀ in

1,2 J-, d a function of time, of the fact by example intensity fluctuations of the light source, asymmetries in the transmission channels electro-optical, and/or the sensitivity of phodétecteurs.

According I⁹ invention₉ the problem is solved in that the constant phasedans the equations (1), (2), is adjusted to- Hf/Q and that the value of the direct current of one of the signals, is set according to that of the other, ±2 o

With can be written the equations (1), (2) in the form:

±_± (t)" k_x L£ + cos (0 + T/4) _7 (3)

±₂ (t)" k₂ L£ + cos (0-VA) _7 (4) electrical signals which can-be described by these equations are, to the phase near, absolutely identical, they exhibit in particular a same value of direct current component. This spring particularly clear, if the measured magnitude is, for example, the line current to alternating current, a sinusoidal function of time therefore, so that 0 (t) can be described by:

0 (t) m 0 cos VA J t (5)

Can then represent the electrical signals® i-^ g1 ^*

±_{1 < t} 2 *^k l,2£^{1 + COS} *^J o + AC_7' (6),

wherein J is Bessel function of the first kind of order zero and the amplitude (0), and AC is the pure alternating quantity, the signals according to (6) have a first DC component that depends only on 2¹ a second additional DC component, which is dependent on the amplitude of the measured quantity. Thus that it is has already indicated, the second component continuous cotirant is different for the signals and Ig, in the first mentioned known solution, in particular when it is zero for " Ÿ " 0.

Therefore only by the invention that it becomes possible, by comparing the actual values of the continuous cotirants DC (i-j _) and DC (Ig) ùes i-^ and ±2* * tin to develop error signal DC (i ^)* _AOE280A2AO> DC (ig), for adjusting the DC equality (i-^)* DC (ig). With such control, * is obtained^k 2"^k '^the P^rem i®^re portion

Mm LEVEL I V

ei top of the problem is thus resolved.

Adjusting ^ " - ^/4 s^s performs no optical parou yen. Starting from the request of the French no. 70,45999 21ο12 · 70, for the value 0 " 0, is adjusted the directions of oscillation of the partial beams, symmetrically with respect to the direction d³ oscillation of the light beam initial total*

To solve also the second part of the problem of the invention, in addition to the solution presented for the value control k-^ "kg" k, is continuously compares, each other, the instantaneous values of the electrical signals, and for the values of the signals above k (" Signalwerte grosser k"), the common value is measured at the time when the two signals are equal. Furthermore, electrically the measured value is divided by cos £Τ^/4*14 V_AOUNI167AO> / 2_e

Equality signals occurs forⁱ l"ⁱ 2

duration

L k, and for i.

Is taken into account, k. cepent This occurs for 0

it is then

±2 χ¾.

that of the case of equality with i,₀

. j. J-

0 O -0 £2 has Tf

The equations (3) " (4) show that i₁ _AOE296A0AO> i₂ "K (1 + cos (£" T^/4)). By measuring Ir/u ±2 st by dividing by 1 + n/2/2, is obtained the valour k, 3rd the way known, subtracted value is k signals i-,₀ k adjusted its function, so that the resulting signals may be described by the equations*

i_li k (t)* cos (0+1f/4) (7)

and

i₂₁ (t)" k cos (0-

(8) As with the solution indicated above, is determined the value of k again, during each successive period signals i^g"^is k can accommodate an variation as a function of time, without impairing the accuracy of the measurement is influenced, Landis than in known solutions k were provided by an external source, k according to the invention is produced from said signal itself₀

Important additional Without means, the method for determining k can be further improved, in that at the time of the equality of the amplitudes not the value of or Ig is measured, but the value i-^ + i₂ , i.e. the sum of the amplitudes of the signals i,₀ 0*0 at the time. The method for determining k is supérietire to the previous one, in particular for higher frequencies of the measured variable"

Other advantages and features of the invention result of the embodiments described below, using the drawings " These show:

in Figure 1, the schematic representation of a measuring device with photoelectron bias modulation and with an electronic device operating, corresponding substantially to the first solutions mentioned, already known;

in Figure 2, execution of the part E of the electronic evaluating means shown in Figure 1, in order to stabilize the amplitudes;

in Figure 3, an extension of the electronic evaluating means according to Figure 2, for measuring the values i ^, i₂ for 0 (t)" 0, to determine the constants and k₂ ; and

in Fig " 4, a possible variant, to replace the Figure 3.

According to Figure 1, a light beam 1 having a field amplitude A passes through a magneto-optical element 3, around which is wound the conductor carrying the current to be measured; it then comes to a blade index 5j and it is split into two partial beams 6 and 7" approximately equal intensity. Before entering the magneto-optical element 3j the light beam is linearly polarized, and as stated with the aid of the crane jib in the circle 2. After passing through the photo-optic element 3j the plane of oscillation of the light beam 1 is turned by an angle 0/2, when a magnetic field is generated in the photo-optic element 3, due to the passage of a current in the conductor. The polarization state of the light beam after passing through the photo-optic element 3" is indicated in the circle 4, where again, the arrow indicates the direction of oscillation.

Index After the blade 3 * the partial beams 6.7 and pass through polarizing filters 8 and 9, whose directions of oscillation are arranged symmetrically with respect to the direction of oscillation, indicated in the circle 2, the initial total light beam, corresponding to the zero value of the line current, respectively and are offset an angle + d * TfVe and- 'Tf' /O with respect to the plane of symmetry"

Therefore, with an amplitude A before the magneto-optical element 5" the photodetector 10 receives the fraction A cos (^ + 0/2/8), and the photodetector 11 receives the fraction At cos (0/2- ΤΤ /8) · Since the photo-current is proportional to the square of the light amplitude, is obtained in the transmission channel 22 the electrical signal

it " k ^ + cos (0+

and in the transmission channel 23, the electrical signal

±2 " ^ Kg * 1 + cos (0-^> Tⁱ / 4) J7"

corresponding to the equations (3) and (4) above < >

The electrical signals i-^ and Ig arrive on the input amplifiers 12,13 and then on the stage E, which comprises the assemblies according to Figures 2 to 4, which will be described further. E The stage is followed multipliers 15 j 16, which can e.g. Stre ring modulators or multipliers four quadrants.

In these multipliers, is multiplied k i^ *^_e cos (0 + Υ by a signal g/4₁ " Cos (^°_Q t-^/4), and Ig '* kg cos (^ - 0/4) by a signal gg " cos (UJ t + ^/4), the carrier frequency.

being generated in a local oscillator 14 and suitably shifted by the phase shifters 17.18 " symmetrical The phase shift is not absolutely necessary. With a suitable phase of the device 21 that still remains to describe, it is sufficient that the components of the carrier frequency are in quadrature phase*

Group For summation at the adder 19 and signals obtained by multiplying, an electrical signal is obtained of the form G (t) "cos (t" 0 (t), that it is easy to then demodulating according to the methods conventional in the device 21, in phase after passage in a limiting amplifier 20, or frequency, after integration 1

η ΐ8b /s

'7.; 7' . lcpc.fVr." 7,77,. ; Ϊ. . '.' · ü. jâgsu.stacaS ;' ja pàasemètredont the

d-3 r- éférencâ is provided by the local oscillator (line year stippled in Figure 1). Phase A diseriaiaateur particularly MSn adapted₅ in the d * obgsfc ime® third requirements of French patent n® 71 ≈ ^ 3554 deposited the 3 february 1971 by the present demanderesseo

The arrangement according to Figure 2 serves for regulating the value of the direct current of one of the signals to be equal to that of the DC component of the other signal, in order to regulate the sensitivity and stabilization of the amplitudes of the signals. To this end, in one of the transmission channels, for example in the channel 23" before the multipliers 13,16 is provided, as an adjusting element, an adjustable-gain amplifier or a multiplier 25, whose setting input is supplied by the output variable of a differential amplifier 24, high amplification, one input of which is connected through the low-pass filter 27,27 *, with the channel 22 for the introduction of the signal which serves as a reference variable, while the other input is connected, through a second low-pass filter 28-, 28 *, with the output of the amplifier 25, to capture the signal ±2 that it set, the low-pass filters have been symbolically represented by HO.

With this arrangement 3 on the terminals 22*, 23' > are obtained stabilized amplitude signals

*^k i £ ^{1 + cos} (0⁺ · ΥΛ ), 7

i*₂ * k, £ 1 4cos( 0 " V/4)_ / o

In Figure 3 "the signals 1 '^ and 1*2 are applied to the input dim comparator 31" from the output of which the voltage changes sign when i i*^ *' g. With this voltage, is controlled monostable multivibrator 32, if the door 41, . inserted into the circuit, is open. During a time equal to its pseudo period, the multivibrator 32 controls the memory, by the circuit sensing and memory 35" of the value of the signal i-^ present on the channel 22, which. at that time, and it is. set forth above, is equal to

i₂ k C 1 * + cos T'A)* 3ε (1+ \ [2/2) o

I-r ds rdgnal output of the sensing circuit and ds memory 33 is applied to the input of the divider 3 ^ " where it is divided by 1+ χ1ζ/2)- the * is arranged such that the value k and by means of mountings adders not shown, can be subtracted from this value signals i ' ^ and i*₂ , to thereby obtain the signals 1 η-, ,±₂₁ according to the equations (7), (8).

the door 41 is controlled by a comparator 42, which compares the actual value of one of the signals, i ^ for example, with the value of k, provided by the divider If i * ^ and therefore i₂ , is smaller than k, the door is closed. Damages to the signal values i ^ " i₂ " Herein for angle values 0 -- n "* n * 1, 2, 3 _AOE280A2AO> o) j, where the invention further i * ^ _AOE296A0AO> i®₂ , are also taken into memory in the holding circuit 33" being in this case i ' i * 1*2*^k (1 > ° - \ J2

The arrangement of Figure 4 is identical to that represented in Figure 3" to the element addit :' canour 3? preset®. With a member, that is the sum of the signals applied to the sensing circuit 1*2 i*2 and memoir ?: 33® Such as divider, is connected to the output of the sensing and memory 33 a voltage divider, composed of resistors 35 and 36, of which one has the value S (1 +t|ao2) and the second has the value H*

The advantage of the circuit arrangement according to Figure 4 appears with the reasoning according to:A the instant of the scanning by the memory circuit 33" read slope ^ and i i®®₂ ast relatively large, especially for the high amplitudes of the function 0 (t). Due to the inherent switching time to the comparator, the control pulse which opens the memory circuit is a delay, so that with a device according to Figure 3, it is possible in the memory that is does not pick up the correct value of the signal scanned, for 0" 0 m 2 i ∩ ρ ₀ However, the sum + i₂ which " in Figure 4 is gripped by the memory circuit, present at the time of scanning a horizontal tangent. A delay of the control pulse is much less in this case.-

TJne attractive alternative, because many more

-economic, of the circuit arrangement according to Figure 4 comprises the removal components 31, 41, 42 and 32, and replace the probe circuit and memory 33 by a peak value detector (53 *)-Such mounting but cannot, in some cases, very rapid tracking the variations in peak values of the function 0 (t), as is required to always discharging first the peak detector, before it is ready for the detection of a new maximum.

The peak detectors and the memory circuits are known, for example by the work of Korn and Korn, Electronic Analog and Hybrid Computers, Mc Graw Hill (1964), pages 378 to 353 PP350 and to 385"