DE1939017A1 - Apparatus for measuring AC signal levels and noise with instant digital reading - Google Patents

Description

PATENT LAWYERS

PROF. DR. DR. J, REITSTÖTTER

DR.-ING, WOLFRAM BUNTE 1939017

O - BODO MÜNCHEN IS, HAYDNSTRASSE 5, FERNRUF <Οβ! 1> S3 47 12

Munich, July 31, 196-9 M / 9754- Lo

I NEW ZEALAND INVENTIONS DEVELOPMENT AUTHORITY

Wellington (New Zealand)

Apparatus for measuring alternating current signal levels and noise with immediate digital reading

The invention relates to digital level meters and in particular! which on a digital decibel meter to measure the strength of the; Level input signal relative "to a prescribed power ^1, typically 1 mW or dB, referred to as the nominal level; is.

j Digital level meters are already known, but they are ! have various drawbacks. For example, there is no option to select the characteristic of the input signal to be measured, i.e. the known system usually provides a quasi mean square root measurement (R.M.S.) in waveform, however, it is often desirable to Average measurements, peak measurements, mean Square root measurements corrected for sine waves or noise

009817 / TOSi

■■■ .. ■. '■■■' ■■. - 2-. . . . ; . ■■; ■ .. ■ -

measurements as well as quasi mean square root measurements. Furthermore, the currently available instruments work with errors, usually correcting them by hand or by Invoice must be made, which is of course disadvantageous.

The invention is based on the object of a digital level meter to create in which the disadvantages mentioned do not occur at all or only to a lesser extent, or at least is a useful choice for the public.

This object is achieved according to the invention by a digital level meter, which is characterized by selection means for adapting the impedance of the instrument to the impedance condition of the line to be measured, switching elements for converting the AC input signal into a DC voltage which is proportional to a selection feature of the input signal, comparison elements firstly for Compare this "DC voltage with a voltage obtained from a discharge resistance Z-capacitance line network and, secondly, to compare this discharge network voltage with one of two DC reference voltages, depending on whether the AC input signal is stronger or weaker than the nominal value, a timing element for measurement the time between the coincidence in the first and that in the second comparison, this time period being proportional to the logarithm of the AC signal voltage with respect to the nominal value, and display and / or memory elements for displaying and / or S. is such that the power level of _t Wechseistromsignalpegels given or obtainable as desired from the storage member ave supplied by the measuring elements memory clock pulses.

Below is. a preferred Ausführungsforra of the invention with reference to the drawing, for example, described in more detail; it ^show::: '_ \' - ■ '- ■■ ---. . ._-. '■■ - _; - ' ^; '. ^; --- ■

0M881 / W6 «

1.0 drawings;

Fig. 1 is a block diagram of a digital according to the invention ... ■ Level meter; . .

FIG. 2 shows a series of waveforms of the signal as shown in FIG a full wave rectifier belonging to the device were corrected;

3 shows diagrams of a mode type selection filter according to the invention with associated waveforms;

Fig. 4 is a graph showing the probability density function in connection with the Gaussian see (white) noise illustrates;

Fig. 5 is a graph showing the discharge characteristics of a R.C. network;

6 is a diagram showing the states of the apparatus relating to of the signal level and the capacitance of the measuring device according to the commissioning, if, the signal level is recorded;

Figure 7 is a sequence diagram of the main sequence counter instructions; ,, and

Fig. 8 is a curve ^ which the erasure of errors by the Device represents.

2.0 Introduction:, -

The embodiment shown in the drawing is a digital one dB meter that is capable of an immediate digital

009 887/1066

"■■■■■ VV - ^: ■■■ '■" ■ ■' ■■ '- -Λ - Λ V ■■ -.- Λ ν

Reading out an AC signal level with respect to a setpoint value to be delivered in decibels; this device can be assembled in one or two ways, either one manually or automatically usable control unit for equipped with an automatic transmission measuring system (ATME) Carrier stations or optionally as portable ones. Measuring instrument. In In the latter case, only minor changes need to be made to the device, which are summarized at the end of the description will.

\ ■

This measuring device was developed in order to measure the values described in several publications of Study Group IV of the C.C.I.T.T. laid down requirements to comply. In particular, the device is designed so that it works extremely accurately, is programmable and that Result

a) in an easily and accurately readable by an operator, as well

b) in one by an electronically controlled Program usable form supplies.

\ -. Λ - '■ ": ■■ -"' ■. V-. ^: ■ --VV V ^:. '■ -; V - "- ^: ■ V-" ■■■■. '. ^: -

3 «Q Table of Contents . V.

3 / facilities

4. Application.

5 * Description -

5.1 General

5 ·? Main sequence counter

5.3 calibration oscillator

5. ^ Signal / calibration and termination unit

5.5 Adjustable preamplifier

5.6 Full wave rectifier

009887 / ΤΟδδ

5.7 Type selection filter

5.8 Filter / integrator unit

5.9 Logarithmic coding device

5.10 Logical coding device ■

5.11 Clock pulse generator

5.12 Recording counter

5.13 Display tube recording OFF / ON, STOF / START control ^

5.14 Type selection unit -

5.15 Signal Area Detector

5.16 Level control counter

6. Oaks'

7. Typical sequence of operations - housing (ATME)

8. Energy supply

9. Portable device

4.0 facilities -

The digital dB meter has those in the following description combined facilities *

4.1 Function types: -

The following can be achieved with the aid of the instrument according to the invention Measurements are taken;

a) Average measurements on each waveform

b) peak measurement on each waveform;

c) Quasi-mean square root measurements on each waveform;

d) Mean measurements - mean square root corrected for sine waves j

e) Noise measurement - calibrated for white and Gaussian noise. . - _ ^<

The first three types of measurements can be made for each waveform will. The fourth type is used for extremely accurate measurement of sine waves (i.e., measurement tone).

Noise measurements are made using the quasi-mean square root method and the result is also averaged over a period of five seconds in order to obtain the estimate to simulate an observer using the counter of a conventional noise meter with a time constant of 200 milliseconds.

k.2 units

The instrument reads for quasi-mean square root, corrected mean square root, and noise measurements dBm off (i.e. the input signal power in terms of ImW), however, could of course read dB with respect to any chosen nominal level.

The instrument is calibrated for peak measurements in such a way that it shows a reading when measuring a sinusoidal wave shape, which is the reading that would appear if the device measured the mean square root value by a number greater than the is equal to the ratio of the peak value to the mean square root value, expressed in dB. In a similar way the device is calibrated for mean value measurements in such a way that it can with

Measuring a sinusoidal waveform shows a reading * which is less than what would appear when measuring the mean square root value by a number that the ratio of the mean value to the mean square root value j expressed in dB, "equals" is the difference between the peak and mean readings therefore correspond the ratio of peak value to mean value, expressed in dB. This difference is correct for non-sinusoidal wave forms.

4.3 Area

Peak measurements) _ ₃₀ . _{j0bls + 10} . _{ö dB}

Mean value measurements)

quasi mean square root measurements)

Corrected mean measurements. ) -3O »O to + 1Q $ O dBm mean square root measurements) '

Noise '-70.0 to -30.0 dBm

The instrument automatically changes the range within these limits in 1OdB steps. If the level of the signal is outside these limits, ⁰ will be displayed on the outside.

4.4 bandwidth

Absolute measurements 300 Hz to 10,000 Hz Relative measurements 200 Hz to 30,000 Hz.

09887/1066

OSSGIfJAL

■■■ ■ '■ .- ■ ■' ■ - - 8 - ^; . '■■■■ - -. ■■■ '. ■' - -

4.5 Overall Accuracy

All types of measurements except noise and peak measurements + 0.1 dB Peak measurements + 0.2 dB

Noise measurements + 0.5 dB

4.6 Change in temperature
+ 10 ° to + 30 ⁰ C.

4.7 Change of energy supply

-48 V.H. Battery: -46 to -52 V.

230 V AC network: -105? the tension

50-5 Hz of the frequency

4.8 Impedance

The instrument is used to measure the level of signals in Transmission lines with symmetrical resistance characteristic impedances of either 600 or 900 ohms. The measurement can be completed or incomplete depending on whether the transmission line is open or not. in the completed state is the instrument's error attenuation better than 30 dB; in the locked state, the bridging loss is less than 0.1 dB.

4.-9 external controls and recording

Outputs I. The result appears on NIXIE tubes in the recording unit 1 with display tube as

0098 8 7/1066

Polarity sign (+ or -)
Tens
Digits
Decimal point
Decimals
II. Two lamps 2 indicate whether

a) the signal is too strong to measure - lamp 2,

b) the signal is too weak to measure - lamp 3. III. A lamp 4 labeled "RESTART" indicates that calibration is required (in the event of a power failure), so that a start switch leading to terminal 5 must be operated.

IV. A neon indicator light (not shown) indicates that the power supply works.

V. The result appearing on the NIXIE tubes will also stored and in BCD form at digital voltage levels between 0 and +3.6 V that are suitable for processing through an automatic embodiment of the Device suitable, made available.

VI. Two electrical outputs are provided to indicate that

a) the signal is in the range and measured became,

b) the signal is out of range of the instrument lies.

Entrances

1. There are electrical inputs to the logical Signals can be routed to carry out the following control processes:

I. Choice of impedance: w-line: completed / not completed

ß '. ■ 600/900 ohm

0Öf8f7 / 1O8t

■ 1938017

■■■. ■, '■■■. ■■ - 10 -' ■■ '; ■ ■'. ■ .... '■

II. Deviation measuring lines ΔΙ / 2, ΔΙ, Δ2, δ4 for plague the nominal level of the signal at which deviation measurements required are; III. Choice of measurement type: Α-line: mean value measurement

Ϊ ^{1 line} : quasi medium qua-. wire root measurements P-line: peak measurement N-line: noise measurement.

ψ If no signal appears on these lines, the level is displayed in the corrected average and mean square root measurement modes.

2. The following switches (not shown) are provided: I. Power Off / On: The mains or the -48V input is connected to the power supply.

II »Measurement calibration: For measurement purposes, by pressing down a calibration signal connected to the CAL terminal with a button, to calibrate the instrument.

III. Another terminal fiff enables inactivation

gain control or range change.

3; To put the device into operation, you either have to use a switch or a remote controlled contact of a Relay connections are made between

I. the lines S and Sl and S2, II. A 170 V output of a power source and the NIXIEr Display tubes in recording thinness 1.

4.10 Waiting for a fee

In use of the automatic embodiment of the apparatus according erf is a device Vorges sh * @ n ₈ old legendary

SADORJQlNAt

-lies the difference between the actually measured signal and a specified value. This set desolate Nominal loss must be between 0.5 and 7.5 dB and one be an integer of half-decibels.

4.11 Calibration

For calibration purposes, as part of the instrument is a at +3 dBm Hz level stable oscillator with an accuracy of up to * 0.01 dB.

The internal calibration takes place automatically whenever the device is switched on and as often as the measurement result crosses 0 dBm. This is set out in more detail below.

Manual calibration is required in the event of a power failure when the power source is being repaired. When this occurs, the "Restart ¹ V lamp k " lights up to draw the operator's attention to this fact.

5.0 application

The basic instrument for level measurement described here is located in a measuring housing and should only be used in a special measuring position. The device can either be used by an operator _f who controls it by depressing the corresponding buttons provided on the housing, or it can be controlled automatically by a program. In the latter case, the NIXIE recording on the display panel of the housing is disregarded and instead the result stored in BCD digital form is used by the program.

00988t / TOSI

The advantage of the NICIE display over analog recording is that it reduces any reading errors caused by the operator. It was found, that such errors account for a significant part of the total number of errors in any level measurement using a Form instruments with an analog recording, e.g. by making parallax errors in the reading or the device is adjusted to the bottles area.

6.0 Description

6.1 General

In simple terms, the instrument is under impedance matching connected to a line and the AC input signal is converted into a DC voltage that corresponds to the peak, is directly proportional to the mean or mean square root of the input signal. The size of this Voltage can then be converted to decibels in the following way will:

It can be shown that the voltage characteristic is opposite the time for a discharge-resistance-Z-capacitance line network (RC line network), which previously up to a known Voltage charged is exponential (or logarithmic) in nature. The magnitude of the DC voltage representing the input signal is determined by the discharge RC line network voltage compared and the time required to achieve a coincidence results in a measurement of the level in decibels. By counting the clock pulses during this period of time with careful selection of the clock pulse frequency, accurate dB measurements can be made are executed. The counting is done using a digital circuit, with the help of which the result

0098 & 7 / 1Q66

■ 1938017

easily stored and / or visually recorded can.

The digital system in the instrument

a) checked the clock pulse count to determine if the Result is in the measuring range of the instrument and whether the range of the level comparator is lower than is that of the instrument. If not, it controls the attenuator settings at the input, the forms the level of the signal in the area of the comparator, provided that the signal level is within the measuring range of the instrument is; ■ -

b) causes the measuring cycle to be continuous in logic Way is repeated.

The following is a description of the individual components:

6.2 Main sequence counter

1. The main sequence counter 5 directs the sequence of operations depending on the type of information it receives. In principle, it consists of a counter composed of four multivibrators (flip-flop circuits), of which six states are displayed using NAND-NOR gates. Monostable multivibrators, which are assigned to a particular state, determine the duration of this state by blocking the clock pulse input to the counter multivibrators for the desired duration of the relevant state. At the end of the pulse of the monostable multivibrator, the clock pulse © can tilt the multivibrator into the next state, actuating the monostable multivibrator assigned to this new state.

9887/1068

2. Sequence of operations

After switching on, the main sequence counter works continuously, giving commands in the form of logical signals to the other units in the following way:

a) Zero position (all logical elements are in a zero position deferred); 100 mS are permissible for this process. Then the instrument goes into the

■ b) Presetting over, during which a stored calibration error ". "- count is entered, whereby 6ö0n seconds are allowed for this presetting process. The instrument can then perform two operations; in one, it integrates the noise level over a period of five seconds before entering a .

c) Measurement state passes over, in which the input signal with the discharge -R. C. line network is compared and the clock pulses are counted during 8mS.

In another sequence, the instrument of the presettings) can lung pass directly to the measurement condition, without the going through noise integration stage. ,

If the result obtained in the measurement state is outside the selected range or outside that of the instrument, the gain is automatically switched on or off Controlled by amplifiers or attenuators *, e.g. attenuators, to bring the signal level to a further selected level bring to. 20 mS are provided for this process, after which the instrument is reset to zero in order to carry out a new cycle in the manner described. "

193S017

However, if the result is in the measurement state in the selected area it is connected to the NIXIE tubes via a gate circuit directed or saved. 6 mS are provided for this process. The instrument then goes back to the zero position and begins a new cycle as described.

3. Summary of the states

State 0 Zero position H 1 Preset Il 2 Measurement Il 3 Gate. It H Reinforcement regulation Il 5 Integrate.

4. Summary of the input and output signals

Inputs: X indicates when the result is too high or too low for the instrument to measure,

RV indicates whether a positive or negative reading is required,

.RV is the logical reverse of RV,

S. the stop-start line shows whether the counter should be turned on or not,

CaI the measuring calibration line indicates that the measuring calibration button is actuated;

OK indicates when the signal is in the correct range,

009887/1066,

CP * "is the pulse train that is used to tilt the meter multivibrator is used and which occurs when the line S is in the logical "start" state,

N indicates that a noise measurement (i.e. no calibration) is required is and therefore the state 5 "Integrate" GewSh-It should be.

Outputs

Ga] / shows the signal / calibration switching unit 6 and the recording counter 1 indicates that calibration is required every time? the instrument is put into operation, as often as the hand elk button is pressed or at every zero crossing of the input level.

χ "" ■ -. ".- ■■ '. ■ -. ■■■ -..""■' - ■ .-," - .--. ■■ - - - CaI is the complement of CaI * " .

Cäli ~ shows the signal / calibration switching unit whether the 3 dBm or the -7 dBm calibration oscillator is required, depending depending on whether the signal is above or below O dBm.

R means zero position (state 0)

R means the complement to the zero position. (State0)

PR means presetting (state 1)

G means gate (state 3)

M means measurement (state 2)

J means gain control (state 4).

00887/1066

- 17 - '■.' ■■ "■

5. Calibration

There are two additional multivibrators in the main sequence counter built in to save the fact that the instrument measures positive or negative values. A NOR gate, which controls the logic signals in the CaIJ line in the counter, determines when calibration is required, as follows:. .

a) When the S _{1 line changes} from stop to start,

b) when the measuring calibration button is pressed,

c) when the strength of the input signal changes, what causes the reading to change from positive to negative or vice versa «

6.3 calibration oscillator

The calibration oscillator 7 consists of a Wrin bridge circuit, wherein the frequency-determining resistors and capacitors are chosen such that they have a frequency of 800 Hz result.

The output level is set to +3 dBm (this value was chosen in order to achieve the greatest error reduction, as will be explained below); the output impedance is set to 600 ohms.

The oscillator is extremely stable and the level Qesamt-Qenauig- is ness in the order of 0.01 dB * This accuracy is achieved by a conventional negative feedback, and comparing! of the output is achieved with an extremely accurate zener voltage , so that any change in the output level changes the drain to the source resistance of an FET on the feedback path. Through an operational amplifier and an in

009887/1066

Series lying FET will yorge- sufficient gain; see to maintain the oscillation.

6.4 Signal / calibration and locking unit

1. This unit6 consists of an attenuator with exactly 600 ohms, 10 dB and an input equalizing transformer (with midpoint capacitor to block DC). The output side of the transformer is branched off to result in the following four w impedance choices.

600 ohms completed 600 ohms not completed 900 ohms completed 900 ohms not completed.

A switchable resistor is included for completed measurements connected to the output side.

The correct impedance tapping point and the necessity of this transformer is determined by a NOR / NAND logic circuit. true, which decodes the instructions contained in the electrical logic signals on the Λ and n input lines /

Relays are provided with contacts » the

a) the input signal, or

b) the + 3 dBm calibration oscillator 7, or.

c) the + 3-dBm calibration oscillator 7 via the 10-dB attenuator 8 (creating a -7dBm «0 oscillator)

connect to the primary winding of the input transformer.

00Θ887 / 1086

These relays are operated by the electrical logic signals on the CaI * and CaI ± input lines. There are additional logic gates are required to ensure that the terminated 600 ohm impedance tap is used during calibration of the transformer is chosen to match the impedance of the calibration oscillator.

2. Entrances

CaI * indicates to the unit whether the 10 dB attenuation * member 8 is to be used or not when calibration is required;

CaI * indicates that calibration is required;

.ir indicates whether a completed or a

unfinished measurement is required;

Ω indicates whether a 600 ohm measurement or a

900 ohm measurement is required;

Balanced AC input for the signal to be measured,; - -----

Input for the +3 ⁱ -dBm-0 ⁱ szill-afo.r.

3. Outputs

The signal produced by the terminating impedance; s_c has it unit has been corrected with regard to impedance, is transmitted to the controllable preamplifier 10, '

887 / iö66

6. Adjustable preamplifier

The adjustable preamplifier 10 is composed of a noise amplifier with a mid-band fixed gain of ¹ IO dB, which is switched into the circuit during the noise measurement, seven in row lying L-damping members, each of which can be switched off, followed by an operational amplifier having an AC Mid-band gain of approximately 40 dB.

The noise amplifier is fuzzy up to 10 kHz and the noise signals with a low level must be in. the dynamic range of the rest of the instrument. This amplifier is controlled by a relay so that whenever the logic level at N * "is a correct level, the amplifier in the circuit is switched on.

The seven L-attenuators used are the following: a 10 dB attenuator and a 20 dB attenuator for the Range adjustment, resulting in a maximum change of 30 dB results;

two 1/2 dB attenuators; a 1 dB attenuator; a 2 dB attenuator and a 4 dB attenuator for deviation measurements a nominal value, giving a range of 7.5 dB in 0.5 dB increments for the nominal value; a 0.912 dB attenuator to achieve the mean square root correction during the measurement according to the averaging method.

The gain of 40 dB is required to prevent the to compensate for the loss caused by the above attenuators and a separate approximately 1-AB attenuator is necessary, " to compensate for the loss of the central volume of the terminating unit.

09867/1066

The attenuators for range adjustment can be one or are switched off and are switched on by logic signals the Y and Z lines controlled by the level meter 11, so that if the strength of the signal is too low or too great is to be read from the logarithmic encoder 12 (which is so designed is that it works with DC voltages corresponding to a dynamic range of 10 dB from 0 to -10 dB, as in Ab ^ Section 6.5 described) processed to werderij, the signal level can be increased or decreased accordingly.

If the 10 dB attenuator is off and the 20 dB attenuator is on, provides an AC signal in the area -ία to 0 dBm a corresponding DC voltage to the logarithmic Encoding device that is in its range.

For example, if the AC input signal in the range -20 to -10 dBm fell, the 10-dB attenuator is on and the 20-dB Därapfungsglied is turned off to the signal to give a Geiaratverefcfirkung of 10 dB. If the change from the romeinganga signal to the range 0 to +10 the Π · 1 ·, the lOwei-Ä ^ fufjiielied "utgiiehfti ^ tt i " or a wtiteie t 10-dB-V "rl4et the signals arise. ;

That equals Prlniip is at the Rtuichiseeaung (in the area • 70 to -30 dBm) applied · Through the noise amplifier an additional gain of 40 dB introduced »

The maintenance of the attenuators that determine the deviation is controlled by logic signals on the lines Δ1 / 2, Al, Δ2 and hk

Beiipitli If Vtrluetabgabetn for a nominal value of 3.5 dB are required, effect the logic signals

009087/1061

BATH

the lines ΔΙ / 2, ΔΙ, Δ2 switch off the attenuators with 1/2, 1 and 2 dB, resulting in an overall gain of Gives 3.5 dB in the system; if an input signal with an absolute level of -4 dB is measured, this shows Instrument -4 + 3.5 = -0.5 dB, which is the required difference between the nominal value -3.5 dB and the actual value, -4 dB corresponds. During the calibration, however, all attenuators in the circuit are switched on again because otherwise an incorrect calibration reading could result.

The switching of the 0.912 dB attenuator is carried out by logical Signals on the Α line are controlled by the type selection unit 13, so that when the average measurement is corrected with middle square root is chosen, the attenuator is switched off and so in the system a gain of 0.912 dB which is equal to the difference between the mean value and the mean square root value for sine waves which remains constant in terms of decibels.

Y Z

42 _s «* Ab» ti £ tmflt§ ») Ut» nti * «ft *» s, ftii * »n

Nominal value d «Ytrluitt iiuut & fen * an de» Deviations are to be measured,

Area unit lines that indicate which Attenuator or which attenuators have to be switched on or off.

indicates that the attenuator is required for the mean measurement with the corrected mean square root »

009887/1068

8ADORIGiNAL

Signal input from the terminating unit,

Cal ^x indicates that calibration is required and is / "used to normalize the deviation attenuators..;; ■ -"

N ^x indicates that a noise measurement is required and is used to turn on the noise amplifier.

3. Exit '-'.

The signal corrected with regard to its level becomes a full-wave rectifier 14 transferred. -

6.6 Full wave rectifier . .

The full wave rectifier l4 consists of two operational amplifiers, which are in series and a precision amplification of the signal to be carried out in two stages.

The first operational amplifier works as a half-wave rectifier with a unity gain, with negative half-waves can be obtained.

The second operational amplifier works as an adder, the half-wave rectified U-signal with the original signal Q combined before rectification so that the one-way rectified negative-going signal has a gain of 2 and the original signal a gain of 1 (unity gain) learns. The result is a fully rectified signal, which appears at the output of the second amplifier, as can be seen from the waveform diagrams in FIG.

The rectified output of the full wave rectifier becomes transferred to the type selection filter 15.

009887/1066

BAD ORiGJWM,

- 2h, -

6.7 Type selection filter

I / The type selection filter 15 consists of passive components which, when connected to one another, form a circuit according to FIG. 3, where d is a perfect diode and E _{0 is} the mean square root voltage. .

The caused by logic signals on the lines P and Λ + Α Proper selection of components ensures that the incoming full wave rectified voltage is for the desired Measurement type - peak, mean, or mean square root measurement - is processed correctly. ·

Referring to Fig. 3, it should be noted that when the resistor Ra without Rb and C would be the output voltage Eout averaged a measure of the mean value of the input waveform represents. Likewise, if Rb and C were present without Ra,. the output waveform averaged a measure of the peak value of the Waveform.

If all the components were in place, the averaged output would be lie somewhere between the peak and the mean. By choosing the right component, the mean square root of a particular waveform, which is also between the peak and mean values can be obtained accurately and the result obtained for a great many other waveforms would be very close to the mean square root. This circuit is called a "quasi mean square root circuit" and has the following properties;

a) Each arm can be switched independently to Carry out mean or peak measurements and quasi mean square root measurements;

b) The source and load resistances of the circuit can during the quasi mean square root measurements and the average measurements in Ra

00988771066

be introduced, whereby the accuracy of the result is improved.

Description of the circuit for quasi mean square root measurements; The time constant of circuit c (Ra + Rb) is chosen so that it is very much greater than the duration of the signal with the lowest frequency likely to be measured, 300 Hz, so that the voltage rise across C is essentially direct current, and if the Diode and Rc not available. it would be the size of the peak voltage of the input waveform.

When steady states are reached, the mean value of the current flowing through Rb is zero and therefore that the voltage across Rb is also zero; that way is the Average voltage at the output of the circuit Eout (medium) practically equal to the DC voltage on capacitor C.

When the diode is permeable and you have a perfect diode assuming that the entrance and the exit are instantaneous, on the same - "potential., - So Eoutj follows Ein -undLtat.Sach-borrowed a fluctuating current flows through Rb, which charges G, however, due to the large time constant, the change in voltage at C is negligible. If the diode is not. Permeable is (caused by the input voltage dropping below the DC voltage at G, modified by Rc and Rb), discharges the capacitor C via Rc and Rb, which are in series, and the instantaneous voltage appearing at the output is equal to dej * DC voltage at G, multiplied by the ratio

Ra -. ■■■ '-' ■■■ -. ■. ■■ '■'. ■■■; ...

Ha ♦ Rb

00988t / t068

Referring to the output voltage waveform diagram of Fig. 3; "S" is defined as that fraction of the Peak voltagej that of the average output voltage is equal to.

It is therefore SE _p = Eo (av) + Vc (the direct voltage at C), and "C" is by definition equal to that fraction of the peak voltage which corresponds to the output voltage when the diode is not permeable, which it is the definition of the quasi mean Square root truncation stress Ec is equal.

So Ec = cEP = Vc χ

ria + rvD

- ^SEP li + Rb

is c = sn or η = -

S.
C.

The ratio - can be determined by varying the resistance values Ra,

Rb can be changed. ■

Being a 'cms) that. desired amount is

009887/1066

For fully rectified sine waves:

κ =

= / T s

& κ = iLÜ

= 1 for C = 0.467 and η = 0.66

or Eo (av) = Ein (r.m.s.) for fully rectified sine waves and it can be shown that Eo (av) for a great many other waveforms is within 0.5 dB of the correct mean Square root value falls if these values of c and η are chosen.

For Gaussian (white) noise

The probability density function, PDF, for white noise can be expressed as

- v ²

p = -rrz: ^xe

(White noise is a statistical term)

where σ = standard deviation or mean square root

the noise voltage
ν = instantaneous noise voltage

It can be shown that when the white noise is ideally fully rectified, there is no change in the mean

009887/1066. _copy

ν. ■■ ■ '■■; ■ '.-;, V- - 28 - _ :

Square root occurs, the probability density puncture however is doubled, i.e.

- / T-1 -

v ²

As before, K =

■ - · ■■. _ for white noise

_ Ec
η σ

= 1 if η = 0.727. - -

and Ec = Ο, 726σ.

However, it was determined to choose η = 0.66 so that the get mean square root of sine waves accurately would,

therefore K = 0.95, which means that there is an error of 5 % or -0.44 dB in the result for white noise. ■

2. Entrances

Signal input from full wave rectifier. P indicates that a peak measurement is required, Λ or A indicates that a mean measurement or an average measurement ^ with the mean square root corrected is required.

Failure to change the state on these two lines means that a quasi mean square root measurement is required.

009887/1066

3. Outputs

The ones for mean, peak and quasi mean square root measurements The corrected signal waveform is transmitted to the filter / integrator unit 16 for averaging.

6.8 Filter / integrator unit

1. The filter / integrator unit 16 consists of an operational amplifier with external components so interconnected that the amplifier can be used as either a Butterworth filter second order (for the purpose of obtaining the mean value, the fundamental or direct current component of the incoming Waveform) or can work as an integrator to the To determine the mean value over a period of time set to five seconds.

The control of the function of this unit either as an integrator or as a filter is carried out by "logic signals on the N ^{x line} which a relay: can respond whose contacts switch on the components, so that if a noise measurement is required, the unit as integrator operates. In this case, the integrating capacitor must be discharged at the completion of the measurement, wherein the control of the discharge is performed by logic signals of the R- (or Nullstellüngs) line. ^l

In all other cases this unit works as a filter.

Similarly, a 6 dB constant current gain is introduced into the filter as a final match to the signal level so that the filter's residual current output is at the correct level (for a given AC input to the instrument)

009887/1086

1938017

> ■■. ■ "■" ■■ ... ^: - - " ^: "" ^: \ - 30 - - ■■ -. ■ - ■ -.;. ■ ■■ -,. .. ■ V.; that logarithm before encoder is given.

The filter is a compromise between a maximum hum attenuation at the lowest frequency (and thus a maximum accuracy of the device) and a minimum switching time (which directly influences the cycle duration of the instrument and is regarded as a time until the end value is reached after a step function change at the input In order to achieve a satisfactory compromise in this regard, it was necessary to use a second order Butterworth filter with a natural time constant of about 10 mS in order to obtain faster separation, thereby reducing the switching time to less than 100 mS can be decreased for a 10 dB instantaneous change in input level, but at the expense of increasing the lowest accurately measurable signal frequency to 300 Hz. ^J.

2. Entrances .

Signal input from the type selection filter

N ^x indicates whether the unit must be operated as a filter or as an integrator,

R indicates the zeroing state of the main sequence counter and becomes the discharge of the integration capacitor used,

3 «outputs

The DC level corresponding to the size of the peak, mean square root or mean value of the AC current, output signal is proportional, is transmitted to the logarithmic encoder 12.

Ö0 9887 / TO86

1938017

6 «9 logarithmic encoder .

1. The logarithmic encoder 12 consists of two comparators X and Y. Each comparator has the property that its initial state through one of two logic levels is determined depending on whether the amount of one input is greater or less than the other. That way changes the output level at voltage levels used to control a logical circle are capable of its state when the Voltage at the input from a higher to a lower voltage, compared to the voltage at the other input, changes, or the other way around.

The following is connected to the comparator X:

a) the DC signal corresponding to the input: from the filter / integrator unit 16, \

b) an R.C. power line previously set to about 5.5V has been charged.

The following is connected to the comparator Y:

a) a reference voltage, one or two specified Assumes values depending on whether the input signal is in a positive or a negative dB range lies. Both reference voltages are supplied by a resistor chain via a temperature-compensated Zener housing branched off,

b) the same R.C. power line as with the comparator X.

2. The working principle is based on the discharge characteristic a previously charged R.C. network, where, if a

009887/1066

1933017,

Measurement is to take place, the capacitor is discharged through the resistor. As can be seen, the characteristic is the Discharge voltage of a logarithmic nature, so that equal time proportions equal voltage change proportions expressed in decibels s'o it is possible to pass the unknown signal through Compare this with the R.C. mains voltage discharging from a known value and measure the time it takes to To achieve the coincidence it is necessary to measure, this time being a direct measurement of the level in decibels, as shown in Fig.4.

3. To measure positive signals, the lower reference voltage Vo is connected to the comparator Y via transistor circuits, under the control of logic signals on lines RV and RV from level regulator counter 11.

By a measurement command coming from the main sequence counter 5 the M line is the charging voltage from the R.C. power line disconnected and discharge begins. For the positive signals, the numerical value of the signal strength increases with the level, the time to be measured must therefore increase as the signal level increases. As can be seen from Fig. 5, the difference between tv and -tviii is the time required. The comparator X shows the Coincidence of Vin and the CR discharge voltage that causes the output to change state when VCR passes Vin passes through and timing begins. Then the comparator shows Y indicates the coincidence of Vo and the CR discharge voltage to end this time measurement. The change of state the outputs of the two comparators, which corresponds to the detection of a level coincidence, is the logic coding circuit 17 fed, where the coming from the clock pulse unit 18 Clock pulses in the period between the change in state of the output of the two comparators - i.e. between tvin and tvo - to be counted.

009867/1066

4. fCR product

A whole count of 800 was chosen for a 10 dB range, where each count corresponds to 1/80 dB, i.e. for a 10 dB range

N _T = 800 = fc (to dB - tlO dB),

where fc = clock pulse _{frequency and t Q} dB - t10 dB means the time it takes for the RC power line to discharge from the voltage corresponding to 10 dB, V10 dB, to the voltage corresponding to 10 dB, Vo dB .

VlO dB - ■■ .-. ■■ ".

^{N = RC 10} ^ e Vo-dB »

da t i RClog _e || ,

where C = capacitance value of the R.C. power line R = resistance value of the R.C. power line

. fRC V10 dB

"Log _1Q e" ¹⁰⁶ IO VO dB

fRC =

A resistor with a value of 1 K ohm was used, a capacitor with the value 0.44 uF and a clock pulse frequency of 1.579 MHz was chosen. .

5. There is a constant error of O, -5 in the encoder; dB built-in,; in order to simplify the calibration process, as this results in proportional errors between 10.5 dB in the coding and analogue stream. can be automatically compensated.

009887/1088

.. ■ ■ - 34 -. ■ ■ ■ ■. ■■■. ■:

6. The high reference voltage is used to measure negative signals V10 connected to the comparator Y, controlled by logic signals on the lines RV and RV.

If the sign of the signals is negative, the numerical value of the result increases as the signal level decreases; the Time to be measured during the discharge of the R.C. power line must therefore increase as the signal decreases. In this case the comparator Y indicates the coincidence of the reference voltage V10 and the R.C. discharge voltage and the corresponding change in level at the output is used by the logic coding device, to start the clock pulse counting.

The comparator X shows the coincidence of the DC input signal and the R.C. discharge voltage, and the corresponding The level change at the output is used to / stop the clock pulse counting. This way the clock pulses counted in the time span tvlQ to tvIN.

In the event that Vin is outside the range Vo to ViO, there are two possibilities;

■■■ ν; : "■■ ■■■■; ' '. " - _; ■; ■. = ■ - ■■■ / ■ .- ■

a) The signal is too strong, in which case the comparator X always changes its state before the comparator Y, there. Vin is greater than VCR, or

b) the signal is too weak, in which case only the comparator Y changes its state, since VCR never reaches Vin.

6. Entrances

DC signal input from filter / integrator 16

00988771066

N indicates that a measurement is required and therefore the R.C. power line must be discharged; ■ "■ - ■ ·:

RV and RV indicate whether the signal is positive or negative is and therefore what reference voltage with the Comparator Y must be connected.

7. 'Outputs '

The outputs of the two comparators are kind of the logical Encoding device connected.

6.10 Logical Coding Device '■

1. The logic coding device 17 consists of two multivibrators Q and W as well as NOR / NAND gates, which identify three states:

a) Ready to measure

b) Measurement in progress - c.) Measurement finished.

When the command to reset from the main sequence counter is received, both multivibrators Q and W is reset and the NAND gate, which controls the clock pulse counting, is switched off by the Q multivibrator held. Both multivibrators are continuously tilted by the clock pulse generator when the line Sl is held at the logic level corresponding to "start".

Signal in the measuring range

If the output of either the comparator X or the

090 87/1066

- 36 V ¹ 938 017

Comparator ^ Y changes according to the beginning of the measurement the multivibrator Q switched on and that the clock pulse series controlling NAND gate activated; the resulting series of clock pulses is transmitted to the recording counter, where the clock pulses are counted.

If the output of the comparator X or Y changes according to the Completion of the measurement changes, change the MuI ti vibrators Q and W their state; the multivibrator Q switches the NAND gate off and thus prevents the transmission of further clock pulses while the multivibrator W is used to hold Q, so that this multivibrator while charging the R.C. power line at the end of the measurement command for the level changes does not react at the comparator outputs.

Now the command to reset is received, the recharge process is finished and the multi! vibrators Q and W are reset so that they are ready for the next measurement.

Signal out of range

k The logical NOR / NAND gates that flip the multivibrators Control Q and W must also be useful in the event that signal Vin is out of range.

// '· "· ■ a) Signal positive and too strong

In this case, the comparator X changes as soon as Vin exceeds VCR , and at the end of the reset the multivibrator Q is toggled to allow the clock pulse to be transmitted to the recording counter. The comparator. Y changes during the measurement command when VCR reaches Vo, whereby the multivibrator Q is set to normal and the multivibrator W is tilted

009887/1066

is going to get a pulse count that is the one for a 10 dB range exceeds, with the sign that the signal is too strong for the measurement in this area.

b) Signal positive and too weak

In this case, the comparator Y changes as soon as Vo increases as a VCR, however, the multivibrator Q can change its state do not change and the pulse series remains locked out with the Indicates that the signal is too weak to be measured in the area concerned. The comparator X does not change. since VCR never reached Vin.

c) Signal negative and too strong *

In this case, the comparator X changes as soon as Vin increases as a VCR, but the multivibrator Q remains in the zero position, The measurement command allows CR to be discharged and the comparator Y changes if V10 exceeds VCR while the. Multivibrator Q cannot change; the pulse series remains locked out with the sign that the signal is to be measured in this Area is too starlc. ■

d) Signal negative and too weak

In this case, the comparator Y changes as soon as V10 becomes greater than VCR, and the multivibrator Q tilts to transfer the series of pulses to the recording counter until the command to reset, whereupon Q is reset to stop the series of pulses. The large one The number of pulses shows that the signal is too low to be increased in this range, using the comparator ' X normally does not change, d§ VCB ''Vin never reached., ". -."'; \ -. '";' - ■ .- ^ ■ - ^: \: ■; ■:.; ^: - ".. '."

α »· ¹ . $ ii / - ι often

-38 .193S017

2 »entrances

Logical inputs from the logarithmic encoder R means zero position (reset) CF * is the clock pulse generator input RV, RT have the same meaning as before.

3. Outputs

CP = clock pulse series, proportional to the signal level in Decibels. -: \ - -.

6 «11 clock pulse ge ne rat or

1. The clock pulse generator 18 consists of a crystal-controlled unstable multivibrator, composed of a * RT CuI 991'Jf integrated circuit. The clock pulse frequency, / J is 1.579 MHz * The output of the multivibrator is determined by ' logic signals on the line Sl blocked, so that Täktimpulse are suppressed when the logic level is on the J Line Sl "Stöp" - corresponds to,; -

2. Entrances

Sl ^s stop / start line

3. Outputs

CP * "s clock pulse series to the main sequence counter, to the logical Coding device and the stop / start control unit ι

009117/1088

■ Ä. ■ - - 39 V> 1 SS 901 7

6 «12 counters

ii ^ The record counter 19 comprises a main counter 20 * _; an error counter 21 and a logical liOR / NAND gate 22 for presetting. The main counter 20 counts the series of pulses coming from the algorithmic encoder 12 during a measurement; the error counter counts the series of pulses when calibration is required. Before each measurement, the circuit for presetting switches the error count from the error counter to the main counter when a presetting command is received from the main sequence counter; this ensures that constant error terms in the input count (which correspond to the error terms introduced before the logarithmic coding process) are suppressed.

The main counter is an " ^II extrapolation" counter and consists of three flip-flops, which divide the incoming clock pulse series by eight, and three digital GuI 9958 serial parallel converters, which are connected in series and indicate the tens, units and decimal counts of the incoming pulse series in BCD form .

If a lock command is received from the main sequence counter 5, the measurement result is in its BCD form in the CuI 9958 housings transferred to temporary storage, the CuI 9959 illuminate the NIXIE tubes via the CuI 9960, which are the BGD outputs convert into decimal form for levels that are suitable for operating NIXIE tubes.

The result is using the device's complementary outputs also as an electrical signal from the CuI 9958 converters available in BCD form

When the main sequence counter receives a reset command becomes the three flip-flops and the three

009887/1086

193S017

CuI 9958 Urasetzer put on hold and ready for the next one counting incoming clock pulse series. The error counter and the Temporary storage remains unchanged. The second decimal place (given by the octal counter) is not on the NIXIE tubes recorded but used

a) for rounding off the result, as later explained

b) for reducing the inevitable Binärzähl- 1 w * pulse fehlers.von

The error counter is a "count down" counter and if from the Main sequence counter 5 receives the CaI * command (which means that calibration is required), forwards the logic Circuit in the recording counter unit 19 the Clock pulse series in the error counter 21 from.

The error counter 21 consists of an octal and a decimal counter (three or four flip-flops), the first and second decimal places in complementary form (count-döwn action) the incoming pulse series counts, which is in direct relation to the size of the level of the built-in calibration oscillator stands. The tens and ones (the integer part) are not recorded. It is believed that the occurring Error * never exceed 0.5 dB; the device is simplified by the introduction of a permanent error of 0.5 dB into the logarithmic coding device 12 (by using reference voltages in the comparator, the + 10.5 dBm and -0.5 dBm are equivalent) introduced in the type of calibration the type of measurement is deleted, along with the proportional errors. An example follows below. The error counter 21 becomes only zeroed when the logic circuit is the

00 9 88771068

Presence of both the Cal * command and the zeroing command of the main sequence counter 5 indicates, which means that a new calibration correction has just been carried out and in the error counter is saved. In fact, the error counter 21 is designed in such a way that it counts to 04 (corresponding to a result of 0.05 dB) is reset, which is used to round off the following to undertake::

Measurement result 3.99 3.51 introduction 0.05 0.05 recording 4.0 3.5 Not recorded 4th 6th

The circuit of the presetting consists of logical NOR / NAND gates, which are under the control of the main sequence counter 5 upcoming preset command

a) the first and second decimal places of a main counter with that of the error counter 21 present the upcoming error count,

.b) preset the ones decade of the main counter to nine, -

c) Put the decade of ten of the main counter on a number which is determined by the range setting and by logical Signals on the Y, Z lines (from the level controller counter) and through logic signals is indicated on the N line, which latter indicates where a noise measurement is being carried out. Details are out Table 1 can be seen.

0098 8771068

1933017

rush ( Measuring range +10 ( dBm 0 ( 0 - -10 -10 - -20 N = O ( -20 - -30 ( -30 - -40 ( -40 - -50 N = 1 -50 - 1 -60 - TABEL

Presetting of the tens on

9 9 0 1

3 4

2. Numerical example for calibration and presetting

A coding error of 4 0.15 dB is introduced into the instrument.

Calibration oscillator level

3 dBm.

Absolute input level to be measured = +6 dBm.

step

Command from the main sequence counter main counter error counting status

Oak trees

1 2 3

CaI

=

Measuring input 1 2

Reset preset

Measurement (logarithm. "Encoder leads +0.5 dB on) + 3 dBm encoded as-A3 dBm

Cal

= 0

Reset preset

previous year
, sung previous Able
sung 00.00 0.05 99.05 0.05 99.05 + 3.65 = 0.05-0 , 65 = 02.70 0.40 00.00 0.40 . -. - - - ■ 99.40
(Error count
in the main
count) 0.40

009887/1086

3 'measurement (+ 6.0OdBm 99th, 40 + 6.65 = 0, '40 encoded as _{n (} - _n (t- \ +6.65 dBm) UD, υο;

Note: The last two numbers are actually counted as 80, not 100.

3. Entrances

= Clock pulse series, proportional to the measurement result j N ^x indicates that a noise measurement is required; CaI * "means that calibration (and therefore the error counter) is required; G. Indicates the gate state; PR indicates the default state; R indicates the state of the zero setting; Y, Z indicate the range setting.

4. Outputs

Measurement result in BCD form for tens, units and decimals -

N ^x complement of line N

8 shows the state of the "8" output line of the ones decade CuI 9958 an; '

D2 shows the state of the "2" output line of the decimal decade CuI 9958.

6.13 Display tube recording, OFF / ON, STOP / START control

1. These units include the NIXIE indicator tubes and associated Control circuit, the limit indicator lamps and the neon light indicating the power supply. A logic and Control circuit ensures that the instrument does not start to work before the output of the power supply is sufficiently

009887/1066

has risen, and that the NIXIE recording is faded out for so long remains until a valid result is obtained.

2. Switch for the power supply

Turning the on / off switch on the power supply connects the mains or the -48 V battery with the built-in energy source.

One connected to the output of the MOV DC power supply R.C.-Netζleitung introduces a delay of about two seconds on, so that there is time to set the logic levels, before the indicator neon light comes on, which means

a) the SI line with a suitable logical. Signal to start is provided and

b) the operator is visually indicated that the power supply is now switched on

■■■■■■ · is. ,

Note: If the operator inadvertently started the instrument before the fluorescent light came on, the above circuit will prevent the instrument from starting.

Similarly, a flip-flop switch K i is controlled by a Schmidt trigger with RC input to delay operation by about two seconds, after two seconds it is brought to the zero position and is used to do the "Restart" Light up lamp 4. This lamp indicates that the Instrument is now ready to start and calibration is required (see below).

In addition, the logic circuit controls the 170 V DC

009887/1086

1938017

Current output, so the NIXIE records at this stage are hidden.

5. Stop / start controls 23 ^;

When the "restart" lamp 4 is lit, the operator can put the instrument into operation, namely

a) either remotely operated by depressing a button in the ATME embodiment, or

b) by pressing a switch, if it is the portable device is acting.

In the first case, the M relay located outside the instrument is actuated. .

connects the line S prepared in advance to the Stop / start line Sl. The consequent change in the logical The level on the line S is indicated by the main sequence counter 5, which states the states described in section 6.2 begins to go through. As already mentioned, it was programmed in such a way that it triggers a calibration measurement every time logic level a change from stop to start on the line Show S.

The relay LM2 connects the 170 V output directly with the NIXIE ^ tube and with the numerical NIXIE tubes above the Circuit controlled by logical NOR / NAND gates, whereby it is ensured that they remain hidden.

The relays LM3 and LM4 connect the signal input of the instrument with the to. measuring transmission line.

009887/1066

; ... ■.; ■■■: - - 46 -

The instrument performs a calibration cycle, but the numerical NIXIE tubes remain hidden. At the end of Calibration measurement, which means the beginning of the first normal measuring process, the flip-flop circuit K is switched on, which the "Restart" lamp goes out.

The logic circuit of the unit indicates the fact that the flip-flop circuit K is switched on and records that the calibration has been completed and the signal is in the correct range - which W is noticeable by the correct logic signal on the OK line - and connects the 170 -V-Energy supply with the NIXIE tubes so that they can now record the result.

4. Power interruption !;

In the event of an interruption in the energy supply, the instrument will of course stop working and the error counter of the The count saved in the recording counter is lost.

When the power supply is restored, the tilt is ί circuit K again turned on and allows the "Restart" lamp light up to indicate that a calibration is required. This can be either

a) by pressing down the calibration button (provided for test purposes), or

b) done by pressing down the button controlling the relay LM. So if the line S is on "Start", changing ^ back to "Stop" and then to "Start" causes the main sequence counter carries out a calibration. If the line S is on "Stop", this is not a problem, because the instrument automatically calibrates itself when the relay LM for a measurement is activated.

00-9887 / 10t *

5. + signal too strong: signal too weak

The two laraps 2 and 3 indicate either

a) that the signal is so strong that it is outside area of the instrument, or

b) that the signal is so weak that it is outside the range of the instrument,

and are indicated by logic signals on lines S ^ g and controlled by the level control counter.

The ί -NIXIE-tube shows either plus or minus, depending after whether the signal is positive or negative and is controlled by logic signals on the RV and RV lines,

6. Entrances

Tens, ones, decimals in BCD form -

CP ^XX is the clock pulse for the flip-flop K; RV, RV indicates whether the signal is positive or negative; OK indicates whether the signal is in the correct range;

CAL indicates calibration is required; CAL ^X is the complement of CAL * ".

7. Outputs -

NIXIE tubes: ί, tens, decimals, ones lamps: signal too weak, S ₁ . _σ signal too strong, "Restart"

. . Decimal point
S: start line
S2: 170 V output
S3:

6.Ί4 Type selection filter 15 and logical unit 13

1. These units comprise logical NOR / NAND gates, which convert the logical signals into signals depending on the desired type of measurement, which

ä) the right attenuator in the adjustable preamplifier look;

b) switch off the correct arm in the type selection filter; C-) make sure that the calibration measurement is always in the

ψ Average and Improved Mean Square

root measurement is carried out, indicated by a logical 0 signal on all type input lines services;

d) the noise amplifier in the adjustable preamplifier for Switch on noise measurements. (The noise amplifier is switched off from the circuit during the calibration * )

2. Singing

:. A: indicates a required average measurement;

Γ: shows a required quasi-mean square. root measurement on; -

indicates a required noise measurement; indicates a required peak measurement; is the reverse of the CAL ^ line and indicates ■■■■ -: - ■ that a calibration is required. ■

3. Outputs

Λ: indicates that the attenuator for the mean and corrected mean square root measurements is off;

009 887/106 $

Λ + A: indicates that either a mean or an average and mean square

zel calibration measurement is required;

N ^x : indicates a noise measurement not to be taken during calibration. ■■ -

6.15 Signal range detector 24

1. This unit comprises two flip-flops, from which four states are derived. NOR / NAND gates branch the clock pulse inputs for the flip-flops from the information arriving on lines U8 and D2 and also the output signals Ü and D, which are dependent on the logic signals on the RV line and the state of the flip-flops.

Lines U8 and D2 of the record counter give one immediate information about whether the signal is too strong, too weak or in range, in the following ways:

As described in Section 6.12, the units decade is in Main counter preset to 9. If this number is not (to O) changes - what discovered by monitoring the line ü8 - there is no clock pulse counting according to section 6.10 1 means that the signal is too weak for the range (if the signal is positive) or that it is suitable for the range is too strong (with negative signal).

If the ones decade changes from 9 to 0 once, it changes the logic signal on line U8; this change says that the signal is now in the range, and it is used to trigger the flip-flops to state 1 (i.e. X9 »5 to YO, 0, since "VCR in the encoder discharges from ViO.5 to ViO dB).

009887/1068

■■■ - ■■■■■. 1933017

".-.- ^: ■■■■■■■■ .. - 50 - ■. '■ .-. - -:;:

If the ones decade changes twice from 9 to 0, the logic signal on line U8 changes again to the same Direction as before and is used to trigger the flip-flops into state 2. This means that the signal is in the range, but only just because it occupies a full 10 dB range (i.e. X9..5 to XO, 0 once, through Χ £, 5 to X £, 0).

At this point line D2 is checked and if it is changes, it means that the signal is out of range; this change is used for the trigger circuits to flip to state 3 (i.e. Z0,0 to Z0,2) '. In this way the instrument has a facility to increase the range by 0.2 dB. If the signal is stronger than 0 dBm, state 3 represents the case in which the signal is too strong for the range and if the signal is below 0 dBm, does this mean that it is too weak for the realm.

The states with reference to Pig. 6 are summarized below: .

For state 0 - no change in U8

For state 1 - U8 changes once

'For state 2 - U8 changes twice

For state 3 "- D2 changes.

The two flip-flops are reset to the state 0 when the measurement command is received from the main sequence counter became.

The NOR / NAND logic gate drops a U signal, which means that more gain is required to make the signal in the range (i.e. that it is too weak), as well as a D signal indicating that the gain has been canceled should be used to bring the signal into the area (i.e.

009887/1066

that it is too strong) and finally an ÖK signal, which indicates that the signal is in the range.

2. inputs; '

U8: BCD-8 output of the one-decade,
D2: BCD-2 output of the decimal decade, M: indicates the measurement status.

3. Outputs ■

Ü: is the complement of U to indicate one for the

Weak signal area;
D: is the complement of D to indicate one for the

Strong Signal Area;
OK: Signal is in range.

6.16 Level control counter ■

The level regulator counter 11 consists of three flip-flops X,

Y and Z connected via NOR / NAND logic gates to to achieve five states and output signals to the lines

Y and Z to supply that of the particular state in which the Flip-flops are offset and depend on the logic levels on lines U and D. The logic levels on lines Y and Z were provided so that the appropriate 10 or 20 dB attenuators from the controllable preamplifier 10 can be turned on or off depending on whether the signal is too strong or too weak for the area.

The CP inputs of the three flip-flops are identical to that of the Main sequence counter 5 incoming J-line connected, so that the Counter only changes its state when an instruction is issued

009887/1068

Pre-gain control is received, i.e. when the signal out of range «. The reset pins of the three Flip-flops are connected to the line Sl so that they only change their state when the line Sl den shows the logical "start" state. If the line Sl in the logical "Stop" state, the flip-flops are in the state. 1 reset in the logic signal level to the Lines Y and Z are supplied, corresponding to the -10 to 0 dBm range for measurements other than noise measurements and accordingly the. —50 to -40 dBm range for noise measurements. In the Indeed, the 20 dB attenuator is switched on and off for state 1 the 10 dB attenuator is switched off, resulting in a voltage loss of 10 dB and a voltage gain of 20 dB in the system arises.

If after the end of a measurement cycle in the range -10 to 0 dBm the signal range detector indicates that the signal is for the range. (D = 0) is too strong, the level control counter 11 changes after receipt of the J command into the state 0, which corresponds to the +10 to 0 dB range, and the logic signals on lines Y, Z change in such a way that both the 10 dB and the 20 dB attenuator can also be switched into the circuit.

If, after a measurement cycle in this +10 to 0 dBm range, the signal range detector indicates that the signal is still too strong for this range, the level control counter changes to state M after receiving the J command. The logic signals on lines Y, Z remain unchanged, but the logic signal on lines S, _T changes Ae and indicates that the signal is too strong to be measured; this change causes the corresponding lamp on the display board to light up.

0 098 87/10 86

If the level of the input signal had to be changed in such a way that that it is in the range of about +10 to O d'Bm. falls what one corresponds to a logical OK signal from the signal range detector, the unit returns to the O state.

If the level of the input signal has dropped instead of getting in the range of -10 to OdBm, this corresponds to a logical Ü = O signal from the signal range detector, whereupon the unit immediately changes to state 1.

The automatic ranking of signals less than -10 dBm is done in a similar way.

The NOR / NAND gates connected to lines Y, Z can determine whether the signal is positive or negative (considering whether to take a noise measurement) and can therefore generate logical RV and RV signals, which are used to display the polarity by means of the ίΝΙΧΙΕ tube as well as the choice of the reference voltage in the logarithmic encoder to control and the CaI signal in the main sequence counter produce. In the example above, the logical signals change the lines RV, RV their state

a) if the unit is from state 1 to state 0 passes; '-

b) when the unit changes from state 4 to state 1 skips to start a calibration cycle.

2 »The sequence diagram is shown in FIG.

3. Inputs "^

U: is the complement of U and indicates whether that Signal is too weak for the area;

009887/1066

■ '■■■ - 5Ά - . ; ·

D: is the complement of D and indicates whether

the signal is too strong for the area; Sl: stop / start control
J: Gate Matching State

N ^x : is the complement of N and indicates whether a noise measurement is required.

4. Outputs

Y, Z: indicate in which range the instrument is working;

RV, RV: indicate whether the input signal is positive or negative; ·.

X: indicates whether the input signal is out of range of the instrument;

S _TC , S _TT : indicate whether the signal is too weak or too strong to be measured by the instrument.

7. 0 calibration

When a signal the device between the forehead or a passes through a similar section of the digital or coding part, the amplitude of this signal is proportional to decibels Signal converted and the proportional errors become fixed errors, which are suppressed by the device, while the fixed errors are made into proportional errors and can be partially suppressed.

With the present instrument, the logarithmic encoding of the measurement result is used to (or similar) section of the instrument Proportional errors as well as those introduced into the coding process to delete fixed errors. (Amplitude proportional errors at the front end are determined by the logarithmic encoding

0 09887/1066

procedure converted into fixed errors). Introduced at the forehead fixed errors and proportional errors in the coding process are compensated for by choosing the level of the calibration oscillator to minimize these errors be decreased when the input voltage is in the middle range and is the same size but opposite at the ends of the range.

Fixed flaws are those that remain constant by yours The amount remains independent of the signal voltage. An example of a fixed error would be the deviation voltage of a ■ Operational amplifier.

Proportional errors are those that affect the input signal voltage are directly proportional. The aging of the components in an amplifier affects e.g. the gain so that the error in the output level depends on the Changes the size of the input level.

Assume that the error voltage introduced by the instrument is "e", composed of proportional _{errors v p} and fixed errors ν '·

• ^e = v _F + v _p

= v _p + KVin,

where Vin is the input signal voltage and K is a proportionality factor is.

The voltage that is used for encoding on the logarithmic encoder is the same

009887 / ΤΟ-ββ

Vin + "e = Vin + ν _π + K Vin

= Vin (1 + K) + V ₁ - ,.

if.

After encoding, the clock pulse count N is proportional the logarithm of the input voltage:

ie Nin = log JVIn (I + K) + v _p -? .

(The "« "sign means" directly proportional to ".)

Likewise, the clock pulse count is N ,, which is obtained when the calibration oscillator is connected to the input,

cal

( ^V cal

The record counter is arranged so that the calibration count is entered into the error counter 21 although the whole count is not stored. The difference between two Counts should be proportional to the logarithm of the difference between the two input voltages, V, and Vin. As a matter of fact

- ^N

cal

Vin

log

1 + k +

Vin
V

VTn

cal

1 + k *

cal

v ", the fixed error component, is designed to be
very much smaller than Vin or V -, is, ie

v _p «Vin

k «1« V

cal

009887/1066

So N. - N _Ί «in cal

log

VIn cal

1 + k - k +

v _p

- cal

where expressions of the second order in the binomial expansion
be disregarded.

Looking closely at the formula, three
be searched.

Pall I. VIn = V,: Disappearance of all, both proportional and fixed, error terms
first order

Case 2. Vin greater than ^V cal ^:

The proportional errors are suppressed and the fixed errors tend to

VW

dwindle, however ψτ- is slightly smaller than

ψ so that the product term under the

^cal unit lies.

Case 3. Vin smaller

as V

cal

The proportional errors cancel each other out and the fixed errors tend to go away, but the product term is slightly larger than the unit "

From this it can be seen that the effect of fixed errors is reduced by choosing a calibration oscillator level at a middle point in the range so that at the ends of the range the are fixed errors of the same magnitude but opposite in sign, as shown in FIG. .. ^ -

A +3 dBm calibration oscillator was chosen for the input level
and a -7 dBm calibration oscillator was used for below 0 dBm

009887/1066

'■:' -: ^ - ■ '.. ■. ■ '- 58 ν. ■ ' : _ ■ /'. '. . ■: ■■ ■■ '■■■;■;..; -

Input level taken. For these areas, calibration oscillators with different polarity are required, because otherwise the errors would be amplified rather than canceled when the oscillator is positive and for one negative dBm range is used and vice versa.

Br Typical sequence of work processes; A.T.M.E.-version ^ form

Example; A level measurement on an unconnected 900 ohm transmission line. The signal level is approx. +6 dBm.

Step! • .Power OFF / ON switch in the ON position:

The DC outputs to the DC converter _s + 3.6V, 170V, -6V, -12V and + 12V reach their final values in a few milliseconds, but the neon indicator lamp in the OFF / ON, START / STOP control lights up after two seconds on to allow a stabilization of the lo- cal ^ level, said line S2 wirdj provided with the correct, intended for the start of the instrument logic signal also this neon lamp shows the operator to ₈ in that the output of the power supply works. At the same time, the toggle switch K in the OFF / ON control unit is brought into the zero position and the "Restart" lamp 4 lights up, which indicates to the operator that the instrument is ready to start.

Step 2; Stop / start circuit

Usually the instrument is turned on Electricity put into operation. Whenever the

&& 9887/1066

housing seated operator to perform a measurement desires, presses the corresponding button, which in this case actuates the relay LM and a mass (or a Voltage with low level) is applied to the Ω input (what indicates that a 900 ohm measurement is to be taken). In this case there is no connection to the line π. The absence of a signal is used by the device as an indication designed so that a completed measurement is desired.

By responding to the relay LM with contacts outside the Instruments achieves the following:

a) Connect the 170V DC power supply to the

t NIXIE tube, the negative sign lights up. The numerical NIXIE tubes remain hidden.

b) Connect the line S2 to the line S, "whereby the resulting change in the line S gives the instrument a Change from stop to start reports which to the following organs transmitted irird:

I. to the clock pulse generator 18, where the clock pulse series CP

can be distributed to different units; II. To the level controller counter 11, where the logical zero setting level on the flip-flops and the latter for a J command to adjust the gain to get prepared; '.

III. to the main sequence counter 5j and logical zeroing level to remove from the flip-flops so that they change their state and after receiving a CP * " Signal to the state 0, RESET (zero position), go can. The flip-flops are in this state 100 mS long held by a suitable one monostable multivibrator in the main sequence counter,

009887/1066

0AD ORIGINAL

1933017; ■

■; ■ ■ - - 6ο - ■■■. . ■. , ■ ■.

which prevents the passage of CP * signals for 100 mS hinders. =, ■.

At the same time, the change in the logical level on the Line S- in the main sequence counter as an instruction to perform a calibration, represented by a change in the logic level on the line CaI *. The -7 dBm calibration oscillator is selected by the CALi line because the first attempt to get the signal in the right area, in the area -10 to 0 dBm is undertaken by the level control counter 11.

c) connecting the instrument to the transmission line. That The signal to be measured, however, cannot get into the instrument, since reversing contacts of relays are controlled by CAL ** "- and CAL lines that lead to the calibration switching unit, the + 3 dBm oscillator via the 10 dB attenuator with the symmetrical Connected to the input of the instrument for calibration purposes. Only when the calibration is finished, the relays are set normally, to connect the signal to be measured to the input of the instrument associate.

The RESET (zeroing) command is transmitted to the logic encoder 17 (to set the MuIt! Vibrators Q and W to zero), as well as to the recording counter (to reset the error counter to 0.05 dB and the main counter to 00.00) and to the filter / integrator 16 (which is not required in this case). The CAL * command is also transmitted to the recording counter 19 in order to prepare the error counter 21 for the clock pulse series CP * and to enable it to be reset.

The time span of 100 mS provided for the zero position enables the relay contacts to be switched over in the Calibration / switching unit and the passing of the -7 dBm signal

00 986 7 /

the terminating unit, if the latter is converted into a terminated 600-ohm impedance for calibration in spite of the logical signals on the lines π and Ω. The signal is then transmitted through the controllable preamplifier 10 when all the deviation adaptation damping elements are switched on in the circuit (this applies during the calibration to ₉ itself, if deviation measurements were desired), the attenuator for the 10 dB range is switched off - and the one for the 20 dB range switched on (corresponding to the range -10 to 0 dBm), and the 0.912 dB attenuator is switched off (since the calibration is carried out as an average and corrected RMS measurement). Then the signal is rectified in the full-wave rectifier lh and transmitted to the type selection filter 15, where (according to a measurement of the average type) only the resistance arm is switched on, and then to the filter / integrator 1β ₉ which works as a filter. A large proportion of the 100 mS duration of the zero setting »state is required by the filter so that its output has enough time to rise to a value which is sufficiently close to the DC end value which must be entered into the logarithmic encoder 12 (es about 800 mS are required). -

Step 3: presetting

At the end of the zero or reset state of 100 mS, the flip-flops in the main sequence counter 5 go into the PRESET (presetting *) state, which is of a very short duration - 600 nanoseconds. The preset command is input to the recording counter 19 to preset the main counter 99 ₈ O5 (determined by logic signals on the lines Y ₉ Z for the decade) Monitoring of the count on lines U8 and D2.

0G9SS7 / 1OSI

_ 52 -

Step 4; Measurement

At the end of the 600 mS preset state, the flip-flops go in the main sequence counter 5 into the measuring state, which is provided long enough to enable the VCR to to discharge properly; the duration is 8 mS. The measurement command is transmitted to the logarithmic encoder 12 to begin discharging the R.C. power line. The level control counter has previously supplied lines RV and RV with logic levels to indicate that the calibration input is negative and therefore the reference voltage corresponding to V10 is required.

The comparator Y will change first if VCR = VClO is rüekzustellen the MUIT! Vibrators Q in the logical encoder 17, 'a may be given _v and the clock pulse train CP * "* the AufzeichnungszShler 19, where the main counter 20 extrapolates (count up ) and the error counter counts down. The comparator X changes when VCR = Vin in order to reset the multivibrator Q and to trigger the multivibrator W, which keeps the former in the zero position and detects any further changes at the outputs of the Makes comparators X and Y independent »

■■■■■■ ^": - -Ζ '''' ■■■■ .:" ■■. ". "■ '- ■ -Ζ

The flip-flops in the signal range detector were once tilted when the logic level of the line U8 has once changed to indicate that the signal ₉ in the region ( »10 to 0 dBm.) Is located; there is a corresponding change in the logic level on the line OK to the main sequence counter 5 "to indicate that the signal is in the range" and to the "OFF / ON" control unit, but since a calibration is in progress, the NIXIE tubes remain hidden "in this way, the main sequence counter able ₈ to gate-state and not the Yerstär" kung settlements state choose au.

■ oo-giif / ioee;

The error count is now stored in the error counter, along with 0.05 dB for rounding purposes.

Step 5: gates

At the end of the measuring state of 8 mS the main sequence counter goes in the gate state. above. The change to the gate state sets the logic level on the line CAL * "normals, which means that the calibration is complete. This State is maintained for 6 mS in order to return to normal enable the relay in the calibration / switching unit (in this case), so that at the end of the gate period the calibration oscillator and the input signal is connected to the instrument. .

The gate command is transmitted to the recording counter 19, where the result in the parallel-to-serial converter of the main counter is entered into the buffer (CuI 9959), The outputs of the buffers are made by CuI 996O converters from the BCD to the decimal form "decrypted and the outputs of these decoders are recorded with the numerical NIXIE recording tied together. However, there is no recording because the NIXIE tubes remain hidden.

The change in the logic level on the CAL * line causes the calibration oscillator to be switched off, as described above, as well as the setting of the flip-flop circuit K in the OFF / ON control unit when the next clock pulse CP * "is received. This causes the" restart "to occur. -Lamp k extinguished and enables the 17Q-V output to be connected to the numerical NIXIE recording as long as the line OK is in the appropriate logic state. The change on the line CAlf also separates the error counter from the incoming one.

ending clock pulse series. The error count, * XY, is up to next calibration measurement saved.

Step 6; Zero position (reset)

At the end of the 6 mS gate state, the main sequence counter returns to its original zero state. The only exception is that the input signal (at approx. +6 dBrn) is present and not the -7 dBm oscillator.

As before, the following processes run in the planned 100 mS away: .

a) the flip-flops in the logic encoder ■ are put on hold and ready for the next measurement;

b) the integrating capacitor in the filter / integrator is discharged, - not required;

c) the main counter is reset to 00.00;

d) the output of the filter / integrator increases up to End value. ^

Step 7: presetting

At the end of the reset state, the main sequence counter goes in the preset state over and - there the signal was in the range -10 dBm to 0 dBm in the last cycle - the As before, the main counter is preset to 99 ".XY, XY is the error count in the error counter Signal range detectors were also reset to monitor the result in the record counter.

Step 8; Measurement

As soon as the DC voltage at the output of the filter, the VIn

009887/1088

1938017-

corresponds to, exceeds VCR, the comparator X changes, but the logic coding device prevents a change of state of the multivibrator Q, If from the main sequence counter the Measurement command is received, the CR power line is discharged and comparator Y changes when V 10 (selected by the logic levels on lines RV, RV) exceeds VCR. It no clock pulse series CP ^ gets into the recording counter, so that the logic level on line U8 does not change. Change the trigger circuits of the signal range detector hence not their condition, which says that the signal is too strong is to be measured »OK on the line to the main sequence counter no change occurs, but line D to the level control counter changes and indicates to the latter that the Reinforcement of the system should be decreased.

Since the line OK shows no change, the main sequence counter must now the gain adjustment state (Adjust Gain) Select. ·

Step 9: gain adjustment

The transition of the multivibrators of the main sequence counter to the gain adjustment state is transmitted via line J to the level control counter, where, in connection with the change in line D, the flip-flops change from state 0 to state '1, which is in the range +10 to 0 dBm is equivalent to. The logic levels on lines Y and Z change to switch the 10 dB attenuator of the adjustable preamplifier into the circuit, which reduces the gain of the system by 10 dB. At the same time, the logic levels on lines RV and change RV " _S which means that the input signal is stronger than 0 dBm.

The change on the lines W and RV used

0099 * 7/1066

a) by the logic circuit in the signal range detectors to determine whether the signal is in the range or not later in the measurement state;

b) through the logarithmic encoder to get the reference voltage change from VlO to Vo "; - ..

c) through the -NIXIE tube recording to go from - to +;

d) through the main sequence counter to initiate another calibration measurement, which becomes necessary because the stored error count was obtained using the -7 dBm calibration oscillator and will therefore not be correct if it is used to compensate for errors in a signal that is above 0 dBm. ^ (The errors would be doubled ₉ instead of being canceled.)

The gain match state lasts 20 mS and provides enough Time for the relays in the calibration / switching unit 6 to to respond to the GAL * "and CAL + commands of the main sequence counter. In this case the +3 dBm calibration oscillator is one and that * Input signal switched off

The change on the line GAI? * "Represents the multivibrator Q in the OFF / ON control unit to zero to the numeric Hide NIXIE tubes *

Step 10; Zero position As for the calibration process described above,

00188771088

Step 11: presetting

As for the calibration process described above. The main counter is still preset to 99 · 05.

Step 12; Measurement ■

The command to start the measurement causes the CR power line to be discharged in the logarithmic encoder. First, the comparator X _{9 changes} when VCR = VIn to initiate the transfer of the clock pulse train into the main and error counters of the record counter. Then, when Vc = VCR, the comparator Y changes to stop the clock pulses. The logic level on the line U8 changes again in order to cause a change on the line OK from the signal range detector to the main sequence counter, with the result that the latter selects the gate state.

The new error count in the error counter is now valid for Signals that are greater than Q dBm, i.e. AB.

Step 13; gate

As for the calibration process described above. The calibration command is removed in the main sequence counter at the beginning of the gate state and the input signal again with the instrument tied together.

The multivibrator Q is set to use the 170 V DC power supply reconnects to the NIXIE recording.

Step IH: Zeroing (resetting) As for the measuring process described above.

009837/1066

_ ■■; ■ ;;. ' . '■' ■■ '..; ■. - 68 - ■.

Step 15: - Presetting

As for the measurement process described above. The new bug count, i.e. 99 AB, is preset in the main counter.

Step * l6; Measurement

As for the measurement process described above; with the exception that Vin now corresponds to the +6 dBm input signal and the Main counter counts up the clock pulses from 99.AB until it reaches 0.60 is when the clock pulse series is interrupted by changing the comparator Y. As before, the signal range detector indicates that it is in range and the line goes OK adjusted accordingly. ;

Step 17-: gate

As described for the measurement process. The result will be +06.0 recorded on the NIXIE tubes and in the caches CuI 9959 stored.

The above cycle is repeated continuously until the operator puts the instrument out of operation by releasing the button on the housing by remote control and thereby the Relay LM switches off.

9.0 Energy supply

9.1 BREATHING version

A DC-DC converter was provided to be used in Sas ΑΤΜΕ housing and meet the energy requirements

009887/1068

1 93 S017

for the following units built into the housing:

a) the digital dB meter

b) three V.F. oscillators.

c) a monitoring amplifier

d) a number of 6V indicator lights. '-

The DC converter is fed by the -50 V public authority battery and allows an input change from -52 V to -46 V. Details of its various outputs are given in Table 2 evident.

The converter switching frequency is high - approximately 50 kHz - and this enables the use of physically small components. A filter circuit was used in the input to prevent hum feedback on the battery to decrease to within 2 millivolts from tip to tip, as in CSP78 "General Clauses Applicable to NZPO Specifications ". All outputs with the exception of the 170-Y» line are regulated. Everyone The regulator has a safety circuit which automatically switches off the regulator transistors and thus separates the charge from the power supply in the event of an overload.

When the power is turned on _s provides an additional circuit a short initial current pulse to overcome the switch and enable the power supply in the position ,, counteract the initial power surge that is required to Glattungskondensatoren charge.

00-8887 / 1088 -

193 8017

9a2 Portable device

The portable embodiment of the digital dB meter has one conventional energy source fed by the mains, with the control, switch-off and start-up circuits already are similar.

The electricity supplied by this energy source is only sufficient borrowed to operate the dB meter.

Output voltage Max .Power demand regulation Tip to j3pj.tze (Volt) (mA) No
Load to "
Full load) Hum at
Full load
(Volt) + 3.6 1500 * -. . - - ■ "- - - -. + 6 * \ 800 +12 400 * -6 ■ 50 -12 150 +170 20th ■ - ■) ■■ - .. ■■■ '■ TABLE 2

The + 6-V »source is not required for the portable version and the power demand from the +! 2 V source is from 400 reduced to 100 mA.

10.0 portable pe res Ge advises

The portable device has largely the same structure but _s has the following variations on

009887/1088

a) A stop / start switch replaces the remote-controlled one Relay LM;

b) to select the impedance settings are switches intended; . . '

c) switches are provided for selecting the type of measurement;

d) the device for performing deviation measurements is not provided, therefore the four deviation attenuators are not present in the controllable preamplifier ₉ and the specified preamplification is reduced from about 40 dB by 7.5 dB to about 32 _r 5 dB;

e) the power source is only used - to provide only enough power for the instrument;

f) the signal input appears on connectors.

009BSl / TOIS

Claims (13)

• 1938017. "V- ■ '■■; - 72 -.' ■■. ■ \ - - '■. ■ - Claims. Apparatus for measuring AC signal levels and noise with instant digital reading, featured by a selection means for adapting the impedance of the Instrument on the impedance condition of the line to be measured, switching elements for converting the AC input signal Firstly, comparison elements into a direct voltage which is proportional to a selection feature of the input signal to compare this DC voltage with one of one Discharge V / resistance / capacitance line network voltage received and second, to compare this discharge line voltage to one of two DC reference voltages as the case may be whether the AC input signal is stronger or weaker than the set point, a timer to measure the time between the coincidence in the first comparison and that in the second comparison, this time period being the logarithm of the AC signal voltage is proportional with respect to the nominal value, and display and / or memory elements for displaying and / or Storing the memory clock pulses supplied by the heating elements in such a way that the power level of the alternating current signal level given or received from the storage element as desired » is cash. 2. Device according to claim 1, characterized by recording means, which is a direct digital readout of the AC signal level deliver in decibels. -- * a 3 »Device according to claim 1, characterized by a calibration member for the automatic compensation of errors in the instrument, consisting of an oscillator whose output level is a whole Number of decibels is turned on in place of the AC input signal at the beginning of a series of measurements is so that when successive input signals in their 009887/1088 ORIGINAL Strength from above the reference input voltage to below the same and vice versa, the clock pulses thus obtained are a measure of the oscillator output level modified by the errors of the instrument result, storage elements in which. these Pulses in complementary form in a memory with a capacity stored in the order of magnitude of 1 dB and added algebraically to the clock pulses corresponding to the desired signal to be measured, so that errors over a range can be suppressed in the order of 1 dB ^ 4. Apparatus according to claim 1 or 2, characterized by a Calibration element for the automatic compensation of errors in the instrument, consisting of an oscillator, whose output level is a is an integer of decibels above a reference level and that is in place of the AC input signal at the beginning of a sequence of measurements is turned on, so that if consecutive Signals in strength greater than the reference input voltage vary to below the same and vice versa, the clock pulses thus obtained are a measure of the error caused by the instrument modified oscillator output level, with this Pulses in complementary form are stored in a memory with a capacity of the order of 1 dB and algebraic are added to the clock pulses corresponding to the desired signal to be measured, so that errors over a Area in the order of 1 dB can be suppressed. 5. The device according to one or more of the preceding claims, characterized in that the selection means has a terminating impedance switch, with selection elements through which the device can be changed so that it results in a selection of impedances _f level matching elements for setting of the nominal level of the signal to which deviation measurements are required, selection elements; through which the 009881/1088 Mean value measurement can be carried out in the instrument, a selection element through which a quasi mean square root measurement can be carried out, a selection element through which a peak measurement can be selected and a selection element, by which a noise measurement can be selected, all under the control of lines to which voltages are applied may or may not be and the presence or absence of a voltage on a particular line indicates the state of the particular, selection member controlled by this particular line certainly. 6. Apparatus according to claim 5 »characterized in that a controllable preamplifier is connected to the terminating impedance unit, which increases the gain with reference to the dB deviation. Adaptation matching lines and to which a logical type selection unit and a type selection filter are connected _s ^{dle are} also connected to a full-wave rectifier, so that the type of conversion of the alternating current signal into a dB voltage can be determined, and that a further filter for filtering. " of the rectified AC signal is provided, said further filter integrating this signal for noise measurement. 7. Apparatus according to claim 5 or 6, characterized in that that to control the controllable preamplifier a level regulator counter is provided with flip-flops. 8. Device according to one or more of the preceding claims, characterized in that circuits and means are provided for the selection and / or control of the sequence of operations of the device, so that a continuous Measurement of the signal level is feasible and these circuits and means also have control members which automatically change the settings of the measuring device when -the signal at the 009881/1066 193S017 '- 75 - ■' * '"' ■■■ Setting in which a previous work sequence was carried out lie outside the measuring range of the device should. 9. Device according to one or more of claims 3'bis 8, characterized in that the two comparison members are built into a logarithmic encoder with which a logical coding device is connected, which has two tilting scarfs * lines and NOR / NAND gates to identify three states, namely a) ready to measure; b) measurement in progress; and c) Measurement completed
having. 10. Device according to one or more of the preceding claims, characterized in that the timing element comprises a clock pulse generator for generating pulses which during the time required for coincidence to be entered in the display and / or storage elements. 11. Device according to one or more of the preceding claims, characterized in that the recording means Rows of NIXIE tubes and associated electronic ones Equipment include. 12. Apparatus according to claim 10 or 11, characterized in that the recording counter is an error counter, a main counter and a preset logic circuit associated with the recording unit. ■ 13. Apparatus according to claim 12, characterized in that the error counter during the zero portion of the cycle of the device not to zero, but to a positive reading 00 98 8 7/1086 1938017 - ■■■ "■■■. ■■■■■■■ ^V ; ■: ■ -.-. 76- ■ .. '■ .-." ■■■■ ". ■. ■■■■■■ ■ -. - is reset so that the next lowest 'recorded Number is rounded up to the next number, 14, device according to claim 12 or 13 »characterized in that that means are provided that count beyond the recorded digit to a further valid digit, so that the otherwise existing binary uncertainty is avoided. Device according to one or more of Claims 12 to Ik _9, characterized in that a signal range detector is connected to the recording counter, the level control counter and the main sequence counter, which controls the level control counter in such a way that * when the incoming signal is outside the measuring range of the device to which this is set, the gain control is operated and the controllable preamplifier and / or attenuators are adjusted and thereby tries to bring the signal level in a range in which it can be measured. 009887/1066 Blank page