Introduction
CiS.01
is an automatic crack detection system to monitor cracks and breaks in
operations with workpieces under mechanical
load. Mainly
CiS.01 is suitable to detect cracks in
shafts during straightening in a straightening press. Special
features that are related to the straightening application are strongly
considered when designing CiS.01 . Every
day thousands of steel shafts are scrapped throughout the automobile industry
as many shafts could not withstand
the high mechanical load during bending/straightening. At
the same time there is the risk of building in defect shafts into gear
boxes when a defect has not been
detected before. Advantages
CiS.01
offers advantages in two fields: 1.
Strong
signal gap in filtering defect workpieces. 2.
Providing
information to optimise the straightening process and to reduce scrap. In
order to achieve this advantages CiS.01 is equipped with these new features:: ·
Optimised
signal detection ( sensor, amplifier, evaluation) ·
Precise
adjustment of measuring to the definite workpiece. ·
Complex
self test function to proof the availability of the system ·
All
measured data are saved for statistic analysis ·
All
data are provided for modern communication standards ( Intranet, e-mail
etc.) ·
Easy
handling through automatic setting and adjusting features . Signal
evaluation
The
sensor converts the signal. It then gets amplified and filtered over frequency.
So signal of low frequency, that often are caused by mechanical noise
can be suppressed. |
|
|
table
1 (Signal Amplitude) |
|
|
Those
signal components that are larger then the noise level adjusted (Noise
Limit: typically 2,5 ‰ of field of
interest) are subject to an energy evaluation. A
large amplitude leads to a quick rise of the signal energy. A little signal
amplitude only provides a slow rise of energy. |
|
|
table
2 (Signal Energy) |
|
|
In
case of crack the signal energy the
curve normally builds up as shown above . The
workpiece is evaluated as cracked when there is overflow in " Limit
Crack ". The
workpiece is evaluated as broken when there is overflow "Limit Break"
. The
limit for the suppression of noise
relates to the signal amplitude. The
limit for crack and break relate to the signal energy. A
crack can already be identified in the lower end of the energy measuring.
The difference between crack and break energy can be set large enough
to enable a clear distinguishing to the
significantly larger break signal. Conditions
for precise recognition: 1.
Good
coupling of sensor to machine 2.
Signal
transmission and amplification with little noise 3.
Parameter
setting of monitor related to the actual workpiece 4.
No
changes of measuring sensitivity of the whole measuring system There
is a number of different influences that may cause difficulties in detecting
cracks. Ignoring these information can lead into wrong results of measuring Sensorial
Aspects
All
piezo electric sensor systems show significant resonance in different
frequencies.. The
resonance picture depends on the structure of the sensor. Also the coupling
of the sensor to the machine and by the machine element itself
influence the resonance. A
reliable judgement about the sensitivity
of the sensor can only be made after
the sensor is mounted to the actual machine. Measuring
sensitivity of the sensor under laboratory
conditions will not allow any prediction to the impulse answer
under production environment. CiS.01
solves this problem by measuring the sensor right after mounting to the
machine. The
impulse is thereby stated by the sensor itself. If
the impulse is stated by a second impulse actor no information on the
coupling of the sensor or the actor
would be available. CiS.01
is using a sensor to create an impulse as well. So the coupling for impulse
and measuring is the same and can be metered. The
piezo element is fitted into a cylindrical shaped alloy steel housing. This
housing is bolted by a central bolt to the machine. Thus
the piezo element receives largest possible service contact to the machine.
There
is no preferred direction for mounting to the machine. The
contact surfaces must be as flat and fine as possible and may not have
any coatings or dirt that could influence the signal transmission. It
is helpful to mount the sensor to the machine as close to the shaft as
possible as the energy is higher the closer
the sensor is to the place of cracking. Signal noise
The
measuring system is completely equipped with differential amplifiers and
transmitters.. CiS.01
allows adjustment of the noise suppression to adjust to individual production
environments . Adjustment
related to the actual workpiece
The
energy that is induced by the
workpiece when cracking depends strongly on the pressure load that had
been put to the workpiece. Small
workpiece diameters show different tensile
strength than larger diameters. Geometric steps may also reduce the tensile
strength of the workpiece. CiS.01
offers to set up to 10 different configurations per workpiece. These
different configurations may be set by different combinations of position of hammer and shaft holders. For
every measuring position signal amplification, noise suppression, rejection
level for crack and break may be set differently ( according to different
shaft design such as geometry, cut ins .. ) The
conditions of measurement could be different for different workpieces
too. Because
of that, CiS.01 has a storage for 50 different workpieces, where the special
parameters applying to the defined measuring positions can be stored. So,
after a set of parameters is defined, a special set for an individual
workpiece can be selected manually by the operator or automatically by
the straightening press. Constant
sensitivity of measurement
With
every new workpiece (a special signal indicates the workpiece transport
to the crack detection unit) a three step selftest is executed by the
detection unit. 1.
By
comparing the special capacities of cables and sensors the unit checks
the electrical connection of cable and sensor to the preamplifier. 2.
By
stimulating the measuring system with a very pricise sine wave, the characteristic
curve of the electrical system is compared to the originally calibrated
curve. 3.
The
actual coupling between sensor and the machine element is checked by the
so called impulse test. The actual measured value of the coupling is compared
with the originally measured and stored value. All
these tests are executed simultaneously with every change of a workpiece
and for up to 8 connected sensors. If
sensors were changed or modifications at the machine were made, the coupling
of sensor and machine will change too. CiS.01
will notify these changes in the coupling. This different coupling means
that there is also a different transfer characteristic of the whole system.
So CiS.01 indicates a fault. After
a simple modification of the machine without changing a sensor it is mostly
possible to bring the coupling into the specified range. If
this is not possible, the new coupling value must be stored. A
changed coupling influences the measuring sensitivity. The
higher the coupling the higher the measuring sensitivity. Data
Security
All
changes of parameters are protected by a 5 level password system. So
only authorized operators can change the system. Measuring
data for statistical evaluation
CiS.01
is equipped with a storage for measured values, that has the capability
to store up to 400,000 workpieces. For
every workpiece CiS.01 stores a serial number, a reference to the type
of workpiece, date and time of manufacturing, the number of strokes and
the results of measuring. And
additionally the measuring result of every stroke (no matter if cracked
or not) is stored. So
by evaluating these data it is possible to recognise changes in the noise
level of the machine, as a result of changes in the quality of the tools
or the system at all. So
you can maintain the machine very early, before getting faults or damaged
workpieces because of wear Changes
made in the preceding manufacturing often become obvious under straightening
stress. The
evaluation of all measured data can help to decide about the advantages
or disadvantages of changes in different steps of the production. For
small amounts of data it is suitable to use a line printer, that can be
connected directly to the crack detection device. It is also possible to send the data to a serial
terminal, if there is a computer connected at the serial RS232 connector
of the CiS.01 unit. To
get more extensive statistical information it is recommended to use an
ethernet connection and a file transfer with FTP protocol. After
an FTP transfer the data is ready for evaluation or storing in an archive
system. Information
about unit status
In
addition to the transfer of statistical data, it is possible to have access
to all the unit parameters. Therefor
CiS.01 is equipped with a built in WEB Server. With
any WEB Browser (it should have java capabilities) you can connect to
the crack detection device and have a look on its settings. CiS.01
can be connected to an Intranet system and configured in a WEB Browser.
Access is protected by a password system. Adjustments
– easy/automatic
The
adjustment of the measuring sensitivity
and the noise suppression for the test functions and the parameter setting
for a single workpiece can be done in a teach procedure. All
settings of the unit can be changed by software. The same CiS.01 unit
is suitable for different types of straightening presses or different numbers of hammers. The
integration of a CiS into a definite machine is normally be done by changing
software parameters. |
CiS.01 © QASS GmbH 2001