Setting the Record Straight:
ADV Data Acquisition Rates and Sampling Volume Size
There has been
some misleading information circulating about newly "improved" specifications of
10-MHz Acoustic Doppler Velocimeters (ADVs). This has been done in an apparent effort to
cause confusion about the true technological advances represented by the SonTek
16-MHz MicroADV. SonTek/YSI would like to address these issues with a full explanation of how
the sampling volume size and data acquisition rates are determined, and to illustrate the
genuine performance advantages of the MicroADV.
Sampling Volume Size: The sampling volume for any type of ADV roughly resembles a
cylinder. The diameter is set by the diameter of the transmit transducer (ADV = 6.0 mm;
MicroADV = 4.5 mm), and the height is set by adjusting the size of the transmit pulse and
receive window through software.
With Doppler techniques, the name of the game is always averaging. The default transmit
pulse length and receive window size are set to allow enough spatial averaging to achieve
our specified 1% accuracy at maximum data output rates (25 Hz for ADVs; 50 Hz for
MicroADVs). The default height of the sampling volume is 7.2 mm for the ADV and 4.5 mm for
the MicroADV. Because the sampling volume is a programmed parameter, it has always been
possible to reduce the sampling volume size by adjusting the size of the transmit pulse
and the receive window length. Indeed, the height of the sampling volume can be decreased
by as much as a factor of six (to a minimum of 1.2 mm for the ADV; 0.75 mm for the
MicroADV). SonTek has been providing this option to anyone who desired it since the very
first ADVs were manufactured. This is hardly a "recent advance".
SonTek has never broadly advertised this capability to adjust the sampling height
because reducing the sampling volume decreases the spatial averaging – effectively
increasing the noise per measurement. Thus, reducing the sampling volume size results in
the instrument no longer being guaranteed 1% accurate at the highest sampling rates. In
many cases, this has proven acceptable for the purposes of our users, but SonTek feels
strongly that anyone who chooses to avail themselves of this technique should be properly
apprised of the possible consequences.
Data Acquisition Rates: All ADVs transmit pulses at very high
rates (typically hundreds of pings per second) to allow averaging of several pings per
sample. Again, this is necessary for the 1% accuracy specification. All ADVLab systems
have long had the capability to sample at faster than advertised rates (25 Hz for ADVs; 50 Hz for MicroADVLabs) through the use of a command line switch in the data acquisition
software. ADVLabs can sample as fast as 100 Hz, while MicroADVLabs can sample as fast as
200 Hz. Again, this is hardly a "recent advance". It has not been broadly
advertised for the same reason presented above. That is, ADVs operating at greater than
their specified maximum data output rates are no longer guaranteed to be 1% accurate,
because fewer pings have gone into the averaging process.
MicroADV Specifications: The MicroADV represents a tremendous improvement in the
spatial and temporal resolution of ADV-type instruments. The MicroADV operates at 16 MHz
in comparison to the 10 MHz of standard ADVs, and it is this difference that results in
the MicroADV’s better resolution.
The MicroADV’s 16-MHz transducers are much smaller than the standard ADV’s
10-MHz transducers, which has the immediate effect of reducing the diameter of the
sampling volume. In addition, the wavelength of 16-MHz sound is 1.6 times smaller than
that of 10-MHz sound, which means the MicroADV can accomplish the same spatial averaging
in a sampling volume that is 1.6 times shorter in comparison to the 10-MHz ADV. The end
result is that the default sampling volume of the MicroADV is about 0.07 cc, roughly
one-third the size of the standard ADV’s 0.2 cc sampling volume.
Another benefit to the 16-MHz frequency is that the higher frequency allows the time
between pulses to be shorter by a factor of 1.6. In addition, 16-MHz sound decays more
rapidly than 10-MHz sound, so there is less need to be concerned about multiple
reflections from nearby boundaries. These factors combine to allow the MicroADV to ping at
much higher rates than the 10-MHz ADV. This means the same number of pings can be sent and
averaged in a much shorter time, with the end result being that the MicroADV is capable of
50-Hz data output rates (twice that of the ADV) while still maintaining the specified 1%
accuracy.
The Capabilities of Reduced Sampling Volume Size and Increased Data Rates are General: All
of the features described above for reducing the sampling volume size and increasing the
data rates of the ADV can be applied to the MicroADV as well. The MicroADV has an absolute
minimum sampling volume that is one-third the size of the absolute minimum sampling volume
of the 10-MHz ADV, yet has the same accuracy. Also, all MicroADV systems have a maximum
data output rate that is twice that of the ADV, yet they have the same accuracy.
Summary: The new improvements being touted for the ADV are in fact old developments
that have long been available for ADVs of any type. In no way should these old
capabilities applied to an ADV be misconstrued as equaling the capabilities of the new
MicroADV. It simply is not the same thing.
SonTek has a long standing policy of being conservative, perhaps overly
conservative, in our specifications. We have not aggressively advertised the
options of reduced sampling volume and increased data rates because we feel
strongly that the use of these features should be done with caution, and with a
full understanding of the potential consequences on the accuracy of the
measurements.
Details about these SonTek/YSI products can be found at:
