EN 61326-1/A1
Störaussendung / Radiation / Emission: Tabelle / table / tableau 4; Klasse / Class /
Classe B.
Störfestigkeit / Immunity / Imunitee: Tabelle / table / tableau A1.
EN 61000-3-2/A14
Oberschwingungsströme / Harmonic current emissions / Émissions de courant
harmonique: Klasse / Class / Classe D.
EN 61000-3-3
Spannungsschwankungen u. Flicker / Voltage fluctuations and flicker /
Fluctuations de tension et du flicker.
Datum /Date /Date Unterschrift / Signature /Signatur
27.03.2001
E. Baumgartner
Technical Manager /Directeur Technique
Instruments
General information regarding the CE marking
HAMEG instruments fulfill the regulations of the EMC directive. The conformity test made by HAMEG is based on the actual generic- and product
standards. In cases where different limit values are applicable, HAMEG applies the severer standard. For emission the limits for residential,
commercial and light industry are applied. Regarding the immunity (susceptibility) the limits for industrial environment have been used.
The measuring- and data lines of the instrument have much influence on emmission and immunity and therefore on meeting the acceptance
limits. For different applications the lines and/or cables used may be different. For measurement operation the following hints and conditions
regarding emission and immunity should be observed:
1. Data cables
For the connection between instruments resp. their interfaces and external devices, (computer, printer etc.) sufficiently screened cables must be
used. Without a special instruction in the manual for a reduced cable length, the maximum cable length of a dataline must be less than 3 meters
and not be used outside buildings. If an interface has several connectors only one connector must have a connection to a cable.
Basically interconnections must have a double screening. For IEEE-bus purposes the double screened cables HZ72S and HZ72L from HAMEG are
suitable.
2. Signal cables
Basically test leads for signal interconnection between test point and instrument should be as short as possible. Without instruction in the manual
for a shorter length, signal lines must be less than 3 meters and not be used outside buildings.
Signal lines must screened (coaxial cable - RG58/U). A proper ground connection is required. In combination with signal generators double
screened cables (RG223/U, RG214/U) must be used.
3. Influence on measuring instruments.
Under the presence of strong high frequency electric or magnetic fields, even with careful setup of the measuring equipment an influence of such
signals is unavoidable.
This will not cause damage or put the instrument out of operation. Small deviations of the measuring value (reading) exceeding the instruments
specifications may result from such conditions in individual cases.
4. RF immunity of oscilloscopes.
4.1 Electromagnetic RF field
The influence of electric and magnetic RF fields may become visible (e.g. RF superimposed), if the field intensity is high. In most cases the
coupling into the oscilloscope takes place via the device under test, mains/line supply, test leads, control cables and/or radiation. The device under
test as well as the oscilloscope may be effected by such fields.
Although the interior of the oscilloscope is screened by the cabinet, direct radiation can occur via the CRT gap. As the bandwidth of each amplifier
stage is higher than the total –3dB bandwidth of the oscilloscope, the influence RF fields of even higher frequencies may be noticeable.
4.2 Electrical fast transients / electrostatic discharge
Electrical fast transient signals (burst) may be coupled into the oscilloscope directly via the mains/line supply, or indirectly via test leads and/or
control cables. Due to the high trigger and input sensitivity of the oscilloscopes, such normally high signals may effect the trigger unit and/or may
become visible on the CRT, which is unavoidable. These effects can also be caused by direct or indirect electrostatic discharge.
HAMEG GmbH
4
Subject to change without notice
Analog Oscilloscope HM1004-3 (100MHz)
Autoset, Save / Recall, Readout / Cursor and RS-232 Interface
Specifications
Vertical Deflection
Operating modes:Channel I or II separate,
Chopper Frequency:approx. 0.5MHz
Sum or difference:from CH I and CH II
Invert:both channels
XY-Mode:via channel I (Y) and channel II(X)
Frequency range:2x DC to 100MHz (-3dB)
Risetime:<3.5ns
Overshoot:≤1%
Deflection coefficients:14 calibrated steps
Automatic (peak to peak): ≤20Hz-200MHz (≥ 0.5div.)
Normal with level control:DC-200MHz (≥0.5div.)
Indicator for trigger action:LED
Slope:positive or negative
Sources:Channel I or II,
ALT. Triggering: CH I/CH II (≥0.8div.)
Coupling:
DC (0 to 200MHz),HF (50kHz – 200MHz),
LF (0 to 1.5kHz), NR(noise reject):0-50MHz (≥ 0.8div.)
Triggering time base B:normal with level
Active TV Sync. Separator:field & line, + / –
External:≥0.3V
Horizontal Deflection
Time base A:22 calibrated steps (±3%)
X-Mag. x10:5ns/div. (±5%)
Holdoff time:variable to approx. 10:1
Time base B:18 calibrated steps (±3%)
Operating modes:A or B, alternate A/B
Bandwidth X-amplifier:0 to 3MHz (-3dB)
Input X-amplifier:via Channel II
Sensitivity:see Ch II
X-Y phase shift:<3° below 220kHz.
Manual(front panel switches)
Auto Set(automatic parameter selection)
Save/Recall: 9 user-defined parameter settings
Readout:Display of parameter settings
Cursor measurement:∆V, ∆t or ∆1/t (frequ.)
Remote control:with built in RS-232 interface
Component Tester
Test voltage:approx. 7V
Test current:approx. 7mA
General Information
CRT:D14-375GH, 8x10div., internal graticule
Acceleration voltage:approx 14kV
Trace rotation:adjustable on front panel
Calibrator:
Line voltage:100-240V AC ±10%, 50/60Hz
Power consumption:approx. 38 Watt at 50Hz
Min./Max. ambient temperature: 0°C...+40°C
Protective system:Safety class I (IEC1010-1)
Weight:approx. 5.9kg. Color: techno-brown
Cabinet:W 285, H 125, D 380 mm
Subject to change without notice.08/00
Channel I and II: alternate or chopped
1mV to 2mV/div.: ±5% (DC – 10MHz (-3dB))
5mV/div. to 20V/div.: ±3% in 1-2-5 sequence,
from 0.5s/div. – 50ns/div. in 1-2-5 sequence
from 20ms/div. to 50ns/div. in 1-2-5 sequence
with variable 2.5:1 up to 50V/div.
line and external.
AC (10Hz – 200MHz),
control and slope selection (0 – 200 MHz)
(0 – 100MHz)
pp
variable 2.5:1 up to 1.25s/div.,with
Operation / Control
(open circuit).
rms
(shorted).
rms
0,2V ±1%, ≈ 1kHz/1MHz (t
<4ns)
r
2 Channels, 1mV – 20V/div, Delay Line, 14kV CRT
Time Base A: 0.5s – 5ns/div., B: 20ms–5ns/div. , 2
nd
Trigger
Triggering DC–200MHz, Automatic Peak to Peak,
Alternate Trigger, Calibrator and Component Tester
This microprocessor controlled oscilloscope has been designed for a wide
multitude of applications in service and industry. For ease of operation the
„Autoset“ function allows for signal related automatic setup of measuring
parameters. On screen alphanumeric readout and cursor functions for
voltage, time and frequency measurement provide extraordinary operational
convenience. Nine different user defined instrument settings can be saved and
recalled without restriction. The built-in RS-232 serial interface allows for
remote controlled operation by a PC .
The outstanding features of the HM1004-3 include two vertical input channels
and the second time base with the ability to magnify, over 1000 times, extremely
small portions of the input signal. The second time base has its own triggering
controls, including level and slope selection,to allow a stable and precisely
referenced display of asynchronous or jittery signal segments. The trigger circuit
is designed to provide reliable triggering to over 200MHz at signal levels as low
as 0.5div.. An active TV Sync Separator for TV-signal tracing ensures accurate
triggering even with noisy signals. Signals are solid and distortion free even at
the upper frequency limit. The built in Y delay line allows for leading edge display
of even low repetition rate signals, supported by the 14kV CRT with its high
intensity. Both instruments are equipped with a built in COMPONENT TESTER.
Because it is so important to be able to trust the accuracy of the display when
viewing pulse or square signals, the HM1004-3 has a built-in switchable
calibrator, which checks the instrument’s transient response characteristics from probe tip to CRT screen. The essential high frequency compensation of
wide band probes can be performed with this calibrator, which features a rise
time of less than 4ns.
The instrument offers the right combination of triggering control, frequency
response, and time base versatility to facilitate measurements in a wide range
of applications - in laboratory as well as in field service use. It is another example
of HAMEG’s dedication to engineering excellence.
Accessories supplied:
Line Cord, Operators Manual on CD-ROM, 2 Probes 10:1
Subject to change without notice
5
General Information
General Information
This oscilloscope is easy to operate. The logical arrangement
of the controls allows anyone to quickly become familiar with
the operation of the instrument, however, experienced users
are also advised to read through these instructions so that all
functions are understood.
Immediately after unpacking, the instrument should be
checked for mechanical damage and loose parts in the
interior. If there is transport damage, the supplier must be
informed immediately. The instrument must then not be put
into operation.
Symbols
ATTENTION - refer to manual
Danger - High voltage
Protective ground (earth) terminal
Use of tilt handle
To view the screen from the best angle, there are three
different positions (C, D, E) for setting up the instrument. If
the instrument is set down on the floor after being carried, the
handle automatically remains in the upright carrying position
(A). In order to place the instrument onto a horizontal surface,
the handle should be turned to the upper side of the oscilloscope (C). For the D position (10° inclination), the handle
should be turned to the opposite direction of the carrying
position until it locks in place automatically underneath the
instrument. For the E position (20° inclination), the handle
should be pulled to release it from the D position and swing
backwards until it locks once more. The handle may also be
set to a position for horizontal carrying by turning it to the
upper side to lock in the B position. At the same time, the
instrument must be lifted, because otherwise the handle will
jump back.
instrument operates according to Safety Class I (threeconductor power cord with protective earthing conductor and
a plug with earthing contact).
The mains/line plug shall only be inserted in a socket outlet
provided with a protective earth contact. The protective
action must not be negated by the use of an extension cord
without a protective conductor.
The mains/line plug must be inserted before connections are made to measuring circuits.
The grounded accessible metal parts (case, sockets, jacks)
and the mains/line supply contacts (line/live, neutral) of the
instrument have been tested against insulation breakdown
with 2200V DC.
Under certain conditions, 50Hz or 60Hz hum voltages can
occur in the measuring circuit due to the interconnection with
other mains/line powered equipment or instruments. This can
be avoided by using an isolation transformer (Safety Class II)
between the mains/line outlet and the power plug of the
device being investigated.
Most cathode-ray tubes develop X-rays. However, the dose
equivalent rate falls far below the maximum permissible
value of 36pA/kg (0.5mR/h).
Whenever it is likely that protection has been impaired, the
instrument shall be made inoperative and be secured against
any unintended operation. The protection is likely to be
impaired if, for example, the instrument
• shows visible damage,
• fails to perform the intended measurements,
• has been subjected to prolonged storage under unfavourable conditions (e.g. in the open or in moist environments),
• has been subject to severe transport stress (e.g. in poor
packaging).
Safety
This instrument has been designed and tested in accordance
with IEC Publication 1010-1 (overvoltage category II, pollu-
tion degree 2), Safety requirements for electrical equipment
for measurement, control, and laboratory use. The CENELEC
regulations EN 61010-1 correspond to this standard. It has left
the factory in a safe condition. This instruction manual contains important information and warnings which have to be
followed by the user to ensure safe operation and to retain the
oscilloscope in a safe condition.
The case, chassis and all measuring terminals are connected
to the protective earth contact of the appliance inlet. The
Intended purpose and operating conditions
This instrument must be used only by qualified experts who
are aware of the risks of electrical measurement.
The instrument is specified for operation in industry, light
industry, commercial and residential environments.
Due to safety reasons the instrument must only be connected
to a properly installed power outlet, containing a protective
earth conductor. The protective earth connection must not be
broken. The power plug must be inserted in the power outlet
while any connection is made to the test device.
The instrument has been designed for indoor use. The
permissible ambient temperature range during operation is
+10°C (+50°F) ... +40°C (+104°F). It may occasionally be
subjected to temperatures between +10°C (+50°F) and -10°C
(+14°F) without degrading its safety. The permissible ambient temperature range for storage or transportation is -40°C
(-40°F) ... +70°C (+158°F). The maximum operating altitude is
up to 2200m (non-operating 15000m). The maximum relative
humidity is up to 80%.
If condensed water exists in the instrument it should be
acclimatized before switching on. In some cases (e.g. extremely cold oscilloscope) two hours should be allowed
before the instrument is put into operation. The instrument
should be kept in a clean and dry room and must not be
operated in explosive, corrosive, dusty, or moist environments. The oscilloscope can be operated in any position, but
the convection cooling must not be impaired. The ventilation
6
Subject to change without notice
General Information
holes may not be covered. For continuous operation the
instrument should be used in the horizontal position, preferably tilted upwards, resting on the tilt handle.
The specifications stating tolerances are only valid if
the instrument has warmed up for 30minutes at an
ambient temperature between +15°C (+59°F) and +30°C
(+86°F). Values without tolerances are typical for an
average instrument.
EMC
This instrument conforms to the European standards regarding the electromagnetic compatibility. The applied standards
are: Generic immunity standard EN50082-2:1995 (for industrial environment) Generic emission standard EN50081-1:1992
( for residential, commercial und light industry environment).
This means that the instrument has been tested to the
highest standards.
Please note that under the influence of strong electromagnetic fields, such signals may be superimposed on
the measured signals.
Under certain conditions this is unavoidable due to the
instrument’s high input sensitivity, high input impedance and
bandwidth. Shielded measuring cables, shielding and earthing
of the device under test may reduce or eliminate those effects.
Warranty
HAMEG warrants to its Customers that the products it
manufactures and sells will be free from defects in materials
and workmanship for a period of 2 years. This warranty shall
not apply to any defect, failure or damage caused by improper
use or inadequate maintenance and care. HAMEG shall not be
obliged to provide service under this warranty to repair
damage resulting from attempts by personnel other than
HAMEG representatives to install, repair, service or modify
these products.
In order to obtain service under this warranty, Customers
must contact and notify the distributor who has sold the
product. Each instrument is subjected to a quality test with 10
hour burn-in before leaving the production. Practically all early
failures are detected by this method. In the case of shipments
by post, rail or carrier the original packing must be used.
Transport damages and damage due to gross negligence are
not covered by the guarantee.
In the case of a complaint, a label should be attached to the
housing of the instrument which describes briefly the faults
observed. If at the same time the name and telephone number
(dialing code and telephone or direct number or department
designation) is stated for possible queries, this helps towards
speeding up the processing of guarantee claims.
ether) can be used to remove greasy dirt. The screen may be
cleaned with water or washing benzine (but not with spirit
(alcohol) or solvents), it must then be wiped with a dry clean
lint-free cloth. Under no circumstances may the cleaning fluid
get into the instrument. The use of other cleaning agents can
attack the plastic and paint surfaces.
Protective Switch Off
This instrument is equipped with a switch mode power supply.
It has both over voltage and overload protection, which will
cause the switch mode supply to limit power consumption to
a minimum. In this case a ticking noise may be heard.
Power supply
The instrument operates on mains/line voltages between
and 240VAC. No means of switching to different input
100V
AC
voltages has therefore been provided.
The power input fuse is externally accessible. The fuse holder
and the 3 pole power connector is an integrated unit. The
power input fuse can be exchanged after the rubber connector
is removed. The fuse holder can be released by lever action
with the aid of a screwdriver. The starting point is a slot located
on contact pin side. The fuse can then be pushed out of the
mounting and replaced.
The fuse holder must be pushed in against the spring pressure
and locked. Use of patched fuses or short circuiting of the fuse
holder is not permissible;
whatsoever for any damage caused as a result, and all warranty
claims become null and void.
Fuse type:
Size 5x20mm; 0.8A, 250V AC fuse;
must meet IEC specification 127,
Sheet III (or DIN 41 662
or DIN 41 571, sheet 3).
Time characteristic: time lag.
Attention!
There is a fuse located inside the instrument within the
switch mode power supply:
Size 5x20mm; 0.8A, 250V AC fuse;
must meet IEC specification 127,
Sheet III (or DIN 41 662
or DIN 41 571, sheet 3).
Time characteristic: fast (F).
The operator must not replace this fuse!
HAMEGHAMEG
HAMEG assumes no liability
HAMEGHAMEG
Maintenance
Various important properties of the oscilloscope should be
carefully checked at certain intervals. Only in this way is it
largely certain that all signals are displayed with the accuracy
on which the technical data are based. Purchase of the
HAMEG scope tester HZ 60, which despite its low price is
highly suitable for tasks of this type, is very much
recommended. The exterior of the oscilloscope should be
cleaned regularly with a dusting brush. Dirt which is difficult
to remove on the casing and handle, the plastic and aluminium
parts, can be removed with a moistened cloth (99% water
+1% mild detergent). Spirit or washing benzine (petroleum
Subject to change without notice
7
Type of signal voltage
Type of signal voltage
The oscilloscopes HM1004-3 and HM1505-3 allow examination of DC voltages and most repetitive signals in the frequency range up to at least 100MHz (-3dB) in case of
HM1004-3 or 150MHz for the HM1505-3.
The vertical amplifiers have been designed for minimum overshoot and therefore permit a true signal display. The display of
sinusoidal signals within the bandwidth limits causes no problems, but an increasing error in measurement due to gain
reduction must be taken into account when measuring high
frequency signals. These errors become noticeable at approx.
40MHz (HM1004-3) or 70MHz (HM1505-3). At approx. 80
MHz (HM1505-3: 110 MHz) the reduction is approx. 10% and
the real voltage value is 11% higher. The gain reduction error
can not be defined exactly as the -3dB bandwidth of the
amplifiers differ between 100MHz and 140MHz (HM1004-3);
and 150MHz and 170MHz (HM1505-3).
For sine wave signals the -6dB limits are approx.
160MHz for the HM1004-3 and 220MHz in the case of
the HM1505-3.
When examining square or pulse type waveforms, attention
must be paid to the harmonic content of such signals. The
repetition frequency (fundamental frequency) of the signal
must therefore be significantly smaller than the upper limit
frequency of the vertical amplifier.
Displaying composite signals can be difficult, especially if they
contain no repetitive higher amplitude content which can be
used for triggering. This is the case with bursts, for instance.
To obtain a well-triggered display in this case, the assistance
of the variable holdoff function or the second timebase may
be required. Television video signals are relatively easy to
trigger using the built-in TV-Sync-Separator (TV).
For optional operation as a DC or AC voltage amplifier, each
vertical amplifier input is provided with a DC/AC switch. DC
coupling should only be used with a series-connected
attenuator probe or at very low frequencies or if the measurement of the DC voltage content of the signal is absolutely
necessary.
When displaying very low frequency pulses, the flat tops may
be sloping with AC coupling of the vertical amplifier (AC limit
frequency approx. 1.6 Hz for 3dB). In this case, DC operation
is preferred, provided the signal voltage is not superimposed
on a too high DC level. Otherwise a capacitor of adequate
capacitance must be connected to the input of the vertical
amplifier with DC coupling. This capacitor must have a sufficiently high breakdown voltage rating. DC coupling is also
recommended for the display of logic and pulse signals,
especially if the pulse duty factor changes constantly. Otherwise the display will move upwards or downwards at each
change. Pure direct voltages can only be measured with DCcoupling.
tions in oscilloscope measurements, the peak-to-peak voltage (Vpp) value is applied. The latter corresponds to the real
potential difference between the most positive and most
negative points of a signal waveform.
If a sinusoidal waveform, displayed on the oscilloscope screen,
is to be converted into an effective (rms) value, the resulting
peak-to-peak value must be divided by 2x√2 = 2.83. Con-
versely, it should be observed that sinusoidal voltages indicated in Vrms (Veff) have 2.83 times the potential difference
in Vpp. The relationship between the different voltage
magnitudes can be seen from the following figure.
The minimum signal voltage which must be applied to the Y input
for a trace of 1div height is 1mVpp (± 5%) when this deflection
coefficient is displayed on the screen (readout) and the vernier
is switched off (VAR-LED dark). However, smaller signals than
this may also be displayed. The deflection coefficients are
indicated in mV/div or V/div (peak-to-peak value).
The magnitude of the applied voltage is ascertained by
multiplying the selected deflection coefficient by the vertical
display height in div. If an attenuator probe x10 is used, a
further multiplication by a factor of 10 is required to ascertain
the correct voltage value.
For exact amplitude measurements, the variable control
(VAR) must be set to its calibrated detent CAL position.
With the variable control activated the deflection sensitivity
can be reduced up to a ratio of 2.5 to 1 (please note “controlsand readout”). Therefore any intermediate value is possible
within the 1-2-5 sequence of the attenuator(s).
With direct connection to the vertical input, signals up
to 400Vpp may be displayed (attenuator set to 20V/
div, variable control to 2.5:1).
With the designations
H = display height in div,
U = signal voltage in Vpp at the vertical input,
D = deflection coefficient in V/div at attenuator switch,
The input coupling is selectable by the AC/DC pushbutton.
The actual setting is displayed in the readout with the “ = “
symbol for DC- and the “ ~ “ symbol for AC coupling.
Amplitude Measurements
In general electrical engineering, alternating voltage data
normally refers to effective values (rms = root-mean-square
value). However, for signal magnitudes and voltage designa-
8
the required value can be calculated from the two given
quantities:
However, these three values are not freely selectable. They
have to be within the following limits (trigger threshold,
accuracy of reading):
Subject to change without notice
H between 0.5 and 8div, if possible 3.2 to 8div,
U between 0.5mVpp and 160Vpp,
D between 1mV/div and 20V/div in 1-2-5 sequence.
Examples:
Set deflection coefficient D = 50mV/div 0.05V/div,
observed display height H = 4.6div,
required voltage U = 0.05x4.6 = 0.23Vpp.
Input voltage U = 5Vpp,
set deflection coefficient D = 1V/div,
required display height H = 5:1 = 5div.
√
Signal voltage U = 230Vrmsx2
(voltage > 160Vpp, with probe 10:1: U = 65.1Vpp),
desired display height H = min. 3.2div, max. 8div,
max. deflection coefficient D = 65.1:3.2 = 20.3V/div,
min. deflection coefficient D = 65.1:8 = 8.1V/div,
adjusted deflection coefficient D = 10V/div.
2 = 651Vpp
Type of signal voltage
Total value of input voltage
The dotted line shows a voltage alternating at zero volt level.
If superimposed on a DC voltage, the addition of the positive
peak and the DC voltage results in the max. voltage (DC +
ACpeak).
The previous examples are related to the crt graticule reading.
The results can also be determined with the aid of the ∆V
cursor measurement (please note “controls and readout”).
The input voltage must not exceed 400V, independent from
the polarity.
If an AC voltage which is superimposed on a DC voltage is
applied, the maximum peak value of both voltages must not
exceed + or -400V. So for AC voltages with a mean value of
zero volt the maximum peak to peak value is 800Vpp.
If attenuator probes with higher limits are used, the
probes limits are valid only if the oscilloscope is set to
DC input coupling.
If DC voltages are applied under AC input coupling conditions
the oscilloscope maximum input voltage value remains 400V.
The attenuator consists of a resistor in the probe and the 1MΩ
input resistor of the oscilloscope, which are disabled by the
AC input coupling capacity when AC coupling is selected. This
also applies to DC voltages with superimposed AC voltages.
It also must be noted that due to the capacitive resistance of
the AC input coupling capacitor, the attenuation ratio depends
on the signal frequency. For sinewave signals with frequencies higher than 40Hz this influence is negligible.
With the above listed exceptions HAMEG 10:1 probes can be
used for DC measurements up to 600V or AC voltages (with
a mean value of zero volt) of 1200Vpp. The 100:1 probe HZ53
allows for 1200V DC or 2400Vpp for AC.
Time Measurements
As a rule, most signals to be displayed are periodically
repeating processes, also called periods. The number of
periods per second is the repetition frequency. Depending on
the timebase setting (TIME/DIV.-knob) indicated by the
readout, one or several signal periods or only a part of a period
can be displayed. The time coefficients are stated in ms/div,µs/div or ns/div. The following examples are related to the
crt graticule reading. The results can also be determined with
the aid of the ∆T and 1/∆T cursor measurement (please note“ Controls and Readout”).
The duration of a signal period or a part of it is determined by
multiplying the relevant time (horizontal distance in div) by the
(calibrated) time coefficient displayed in the readout .
Uncalibrated, the timebase speed can be reduced until a
maximum factor of 2.5 is reached. Therefore any intermediate value is possible within the 1-2-5 sequence.
With the designations
L = displayed wave length in div of one period,
T = time in seconds for one period,
F = recurrence frequency in Hz of the signal,
Tc = time coefficient in ms, µs or ns/div and the relation
F = 1/T, the following equations can be stated:
It should be noted that its AC peak value is derated at higher
frequencies. If a normal x10 probe is used to measure high
voltages there is the risk that the compensation trimmer
bridging the attenuator series resistor will break down causing damage to the input of the oscilloscope.
However, if for example only the residual ripple of a high
voltage is to be displayed on the oscilloscope, a normal x10
probe is sufficient. In this case, an appropriate high voltage
capacitor (approx. 22-68nF) must be connected in series with
the input tip of the probe. With Y-POS. control (input coupling
to GD) it is possible to use a horizontal graticule line as
reference line for ground potential before the measurement. It can lie below or above the horizontal central line
according to whether positive and/or negative deviations
from the ground potential are to be measured.
Subject to change without notice
However, these four values are not freely selectable. They
have to be within the following limits:
Lbetween 0.2 and 10div, if possible 4 to 10div,
Tbetween 5ns and 5s,
Fbetween 0.5Hz and 100MHz,
Tcbetween 50ns/div and 500ms/div in 1-2-5 sequence
(with X-MAG. (x10) inactive), and
Tcbetween 5ns/div and 50ms/div in 1-2-5 sequence
(with X-MAG. (x10) active).
Examples:
Displayed wavelength L = 7div,
set time coefficient Tc = 100ns/div,
required period T = 7x100x10-9 = 0.7µs
required rec. freq. F = 1:(0.7x10-6) = 1.428MHz.
Signal period T = 1s,
9
Type of signal voltage
set time coefficient Tc = 0.2s/div,
required wavelength L = 1:0.2 = 5div.
Displayed ripple wavelength L = 1div,
set time coefficient Tc = 10ms/div,
required ripple freq. F = 1:(1x10x10-3) = 100Hz.
TV-Line frequency F = 15625Hz,
set time coefficient Tc = 10µs/div,
required wavelength L = 1:(15 625x10-5) = 6.4div.
Sine wavelength L = min. 4div, max. 10div,
Frequency F = 1kHz,
max. time coefficient Tc = 1:(4x103) = 0.25ms/div,
min. time coefficient Tc = 1:(10x103) = 0.1ms/div,
set time coefficient Tc = 0.2ms/div,
required wavelength L = 1:(103x0.2x10-3) = 5div.
Displayed wavelength L = 0.8div,
set time coefficient Tc = 0.5µs/div,
pressed X-MAG. (x10) button: Tc = 0.05µs/div,
required rec. freq. F = 1:(0.8x0.05x10-6) = 25MHz,
required period T = 1:(25x106) = 40ns.
If the time is relatively short as compared with the complete
signal period, an expanded time scale should always be applied
(X-MAG. (x10) active). In this case, the time interval of interest
can be shifted to the screen center using the X-POS. control.
When investigating pulse or square waveforms, the critical
feature is the risetime of the voltage step. To ensure that
transients, ramp-offs, and bandwidth limits do not unduly
influence the measuring accuracy, the risetime is generally
measured between 10% and 90% of the vertical pulse height.
For measurement, adjust the Y deflection coefficient using its
variable function (uncalibrated) together with the Y-POS.
control so that the pulse height is precisely aligned with the
0% and 100% lines of the internal graticule. The 10% and
90% points of the signal will now coincide with the 10% and
90% graticule lines. The risetime is given by the product of the
horizontal distance in div between these two coincident
points and the calibrated time coefficient setting. The fall time
of a pulse can also be measured by using this method.
The following figure shows correct positioning of the oscilloscope trace for accurate risetime measurement.
the risetime of the probe (e.g. = 2ns). If t
34ns, then t
calculation is unnecessary.
can be taken as the risetime of the pulse, and
tot
is greater than
tot
Calculation of the example in the figure above results in a
signal risetime
2
t
r
- 3,52 - 22 = 6,9ns
= √ 8
The measurement of the rise or fall time is not limited to the
trace dimensions shown in the above diagram. It is only
particularly simple in this way. In principle it is possible to
measure in any display position and at any signal amplitude.
It is only important that the full height of the signal edge of
interest is visible in its full length at not too great steepness
and that the horizontal distance at 10% and 90% of the
amplitude is measured. If the edge shows rounding or overshooting, the 100% should not be related to the peak values
but to the mean pulse heights. Breaks or peaks (glitches) next
to the edge are also not taken into account. With very severe
transient distortions, the rise and fall time measurement has
little meaning. For amplifiers with approximately constant
group delay (therefore good pulse transmission performance)
the following numerical relationship between rise time tr (in
ns) and bandwidth B (in MHz) applies:
Connection of Test Signal
In most cases briefly depressing the AUTO SET causes a
useful signal related instrument setting. The following explanations refer to special applications and/or signals, demanding a manual instrument setting. The description of the
controls is explained in the section “controls and readout”.
Caution:
When connecting unknown signals to the oscilloscope
input, always use automatic triggering and set the
input coupling switch to AC (readout). The attenuator
should initially be set to 20V/div.
With a time coefficient of 5ns/div (X x10 magnification active),
the example shown in the above figure results in a total
measured risetime of
t
= 1.6div x 5ns/div = 8ns
tot
When very fast risetimes are being measured, the risetimes
of the oscilloscope amplifier and of the attenuator probe has
to be deducted from the measured time value. The risetime
of the signal can be calculated using the following formula.
2
2
= √ t
t
r
In this t
of the oscilloscope amplifier (HM1004-3 approx. 3.5ns) and t
tot
- t
tot
osc
is the total measured risetime, t
- t
2
p
is the risetime
osc
10
Sometimes the trace will disappear after an input signal has
been applied. Then a higher deflection coefficient (lower input
sensitivity) must be chosen until the vertical signal height is
only 3-8div. With a signal amplitude greater than 160Vpp and
the deflection coefficient (VOLTS/DIV.) in calibrated condition, an attenuator probe must be inserted before the vertical
input. If, after applying the signal, the trace is nearly blanked,
the period of the signal is probably substantially longer than
the set time deflection coefficient (TIME/DIV.). It should be
switched to an adequately larger time coefficient.
The signal to be displayed can be connected directly to the Yinput of the oscilloscope with a shielded test cable such as
HZ32 or HZ34, or reduced through a x10 or x100 attenuator
probe. The use of test cables with high impedance circuits is
only recommended for relatively low frequencies (up to
approx. 50kHz). For higher frequencies, the signal source
must be of low impedance, i.e. matched to the characteristic
resistance of the cable (as a rule 50Ω). Especially when
transmitting square and pulse signals, a resistor equal to the
characteristic impedance of the cable must also be connected
across the cable directly at the Y-input of the oscilloscope.
When using a 50Ω cable such as the HZ34, a 50Ω through
termination type HZ22 is available from HAMEG. When
transmitting square signals with short rise times, transient
phenomena on the edges and top of the signal may become
p
Subject to change without notice
Controls and Readout
visible if the correct termination is not used. A terminating
resistance is sometimes recommended with sine signals as
well. Certain amplifiers, generators or their attenuators maintain the nominal output voltage independent of frequency only
if their connection cable is terminated with the prescribed
resistance. Here it must be noted that the terminating resistor
HZ22 will only dissipate a maximum of 2Watts. This power is
reached with 10V
or x100 attenuator probe is used, no termination is necessary.
In this case, the connecting cable is matched directly to the high
impedance input of the oscilloscope. When using attenuators
probes, even high internal impedance sources are only slightly
loaded (approx. 10MΩ II 12pF or 100MΩ II 5pF with HZ53).
Therefore, if the voltage loss due to the attenuation of the
probe can be compensated by a higher amplitude setting, the
probe should always be used. The series impedance of the
probe provides a certain amount of protection for the input of
the vertical amplifier. Because of their separate manufacture,
all attenuator probes are only partially compensated, therefore
accurate compensation must be performed on the oscilloscope (see Probe compensation ).
Standard attenuator probes on the oscilloscope normally
reduce its bandwidth and increase the rise time. In all cases
where the oscilloscope bandwidth must be fully utilized (e.g.
for pulses with steep edges) we strongly advise using the
probes HZ51 (x10) HZ52 (x10 HF) and HZ54 (x1 and x10). This
can save the purchase of an oscilloscope with larger bandwidth.
The probes mentioned have a HF-calibration in addition to low
frequency calibration adjustment. Thus a group delay correction to the upper limit frequency of the oscilloscope is possible
with the aid of an 1MHz calibrator, e.g. HZ60.
or at 28.3Vpp with sine signal. If a x10
rms
adapter, should be used. In this way ground and matching
problems are eliminated. Hum or interference appearing in
the measuring circuit (especially when a small deflection
coefficient is used) is possibly caused by multiple grounding
because equalizing currents can flow in the shielding of the
test cables (voltage drop between the protective conductor
connections, caused by external equipment connected to the
mains/line, e.g. signal generators with interference protection capacitors).
Controls and Readout
The following description assumes that the instrument is not set to “COMPONENT TESTER” mode.
If the instrument is switched on, all important settings are
displayed in the readout. The LED´s located on the front panel
assist operation and indicate additional information. Incorrect
operation and the electrical end positions of control knobs are
indicated by a warning beep.
Except for the power pushbutton (POWER), the calibrator
frequency pushbutton (CAL. 1kHz/1MHz), the focus control(FOCUS) and the trace rotation control (TR) all other controls
are electronically selected. All other functions and their settings can therefore be remote controlled and stored.
The front panel is subdivided into sections.
On the top, immediately to the right of the CRT screen,
the following controls and LED indicators are placed:
In fact the bandwidth and rise time of the oscilloscope are not
noticably changed with these probe types and the waveform
reproduction fidelity can even be improved because the probe
can be matched to the oscilloscopes individual pulse response.
If a x10 or x100 attenuator probe is used, DC input
coupling must always be used at voltages above 400V.
With AC coupling of low frequency signals, the attenuation is no longer independent of frequency, pulses
can show pulse tilts. Direct voltages are suppressed
but load the oscilloscope input coupling capacitor
concerned. Its voltage rating is max. 400 V (DC + peak
AC). DC input coupling is therefore of quite special
importance with a x100 attenuation probe which usually has a voltage rating of max. 1200 V (DC + peak AC).
A capacitor of corresponding capacitance and voltage
rating may be connected in series with the attenuator
probe input for blocking DC voltage (e.g. for hum
voltage measurement).
With all attenuator probes, the maximum AC input voltage
must be derated with frequency usually above 20kHz. Therefore the derating curve of the attenuator probe type concerned must be taken into account.
The selection of the ground point on the test object is
important when displaying small signal voltages. It should
always be as close as possible to the measuring point. If this
is not done, serious signal distortion may result from spurious
currents through the ground leads or chassis parts. The
ground leads on attenuator probes are also particularly critical.
They should be as short and thick as possible. When the
attenuator probe is connected to a BNC-socket, a BNC-
(1) POWER - Pushbutton and symbols for ON (I) and OFF
(O).
After the oscilloscope is switched on, all LEDs are lit and
an automated instrument test is performed. During this
time the HAMEG logo and the software version are
displayed on the screen. After the internal test is completed succesfully, the overlay is switched off and the
normal operation mode is present. Then the last used
settings become activated and one LED indicates the ON
condition.
Some mode functions can be modified (SETUP) and/or
automated adjustment procedures (CALIBRATE) can be
called if the “MAIN MENU” is present. For further
information please note “MENU”.
(2) AUTO SET -
Briefly depressing this pushbutton (please note “AUTO
SET”) automatically selects Yt mode. The instrument is
set to the last used Yt mode setting (CH I, CH II or DUAL).
Even if alternating timebase mode or B timebase mode
was active before, the instrument is switched automatically to A timebase mode. Please note “AUTO SET”.
Automatic CURSOR supported voltage measurement
If CURSOR voltage measurement is present, the CURSOR lines are automatically set to the positive and
negative peak value of the signal. The accuracy of this
function decreases with higher frequencies and is also
influenced by the signal‘s pulse duty factor.
Subject to change without notice
11
Loading...
+ 25 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.