I2773-4.
4
en
209
Somat eDAQ
lite
14 Data Synchronization
This section describes the limitations on the synchronization of the data samples.
eDAQ
lite
data acquisition synchronization across channels is accomplished by using
a single master clock source that drives the data acquisition hardware. The term lag
indicates that, in a Time History plot, the data appears later than it should while the
term lead indicates that the data appears earlier than it should.
NOTE
Unless otherwise noted, the following discussion and numerical examples assume a
100000 Hz master clock rate.
14.1
Data Synchronization Characterization Method
To characterize eDAQ
lite
data synchronization, a ±5000-millivolt triangle function
generator waveform is fed in parallel into all channels to be characterized. The
frequency of the waveform is set at the sample rate divided by 1000 to yield 1000
sample points per cycle. For each reversal, all data samples that fall between ±2000
millivolts are least squares fit to provide a very accurate measurement of the zero
crossing time. The differences in these zero crossing times from one channel to the
next represent the data skew from one channel to the next. For each test run, the data
skew on at least 200 consecutive reversals is measured and then averaged. At least
3 test runs are performed and the average data skew over the set of test runs is the
characterized data skew value.
14.2
Analog Channel Synchronization
The ELHLS and ELBRG channels all employ pre-start periods to compensate for their
analog guard filters. In addition to the guard filter skew, there are some other
secondary factors that influence data synchronization, such as A/D converter
conversion time and transport delays through gain amplifiers.
NOTE
This discussion assumes no digital filtering. Ideally, linear phase digital filters do not
result in phase shifts. For the ELHLS and ELBRG channels, however, the linear phase
filters for sample rates at or below 10000 Hz result in a five microsecond lead data
skew. The Butterworth digital filters are designed to match their analog equivalents
and, therefore, these filters do generate significant phase shifts which, in turn,
significantly affect synchronization with other transducer channels.
Following is a table that contains actual data skew characterization test results (in
microseconds) for one eDAQ
lite
stack. The data in this table is consistent with the
data skew times discussed in this section. The first channel on the first ELHLS layer
was arbitrarily used as the data sync time reference channel. The test covered the
channels on two ELHLS layers and the channels on one ELBRG layer. The data sync
was characterized using three different sample rates for both the 100 kHz and the
98.304 kHz master sample rates.
HBM: public
Summary of Contents for eDAQlite
Page 1: ...User Manual English Deutsch Espanol I2773 4 4 en HBM public Somat eDAQlite with TCE Software ...
Page 2: ...Somat eDAQlite 2 I2773 4 4 en HBM public ...
Page 16: ...Somat eDAQlite 16 I2773 4 4 en HBM public ...
Page 22: ...Somat eDAQlite 22 I2773 4 4 en HBM public ...
Page 88: ...Somat eDAQlite 88 I2773 4 4 en HBM public ...
Page 106: ...Somat eDAQlite 106 I2773 4 4 en HBM public ...
Page 128: ...Somat eDAQlite 128 I2773 4 4 en HBM public ...
Page 190: ...Somat eDAQlite 190 I2773 4 4 en HBM public ...
Page 202: ...Somat eDAQlite 202 I2773 4 4 en HBM public ...
Page 222: ...Somat eDAQlite 222 I2773 4 4 en HBM public ...
Page 224: ...Somat eDAQlite 224 I2773 4 4 en HBM public ...