High pressure Aquadopps
Recent results, for example from the Woods Hole Ultramoor program (in the Sargasso Sea, near Bermuda), are showing that the high pressure Aquadopps are working in deep water where scattering is limited. However, since the Ultramoor Aquadopp ws delivered, there have been futher improvements. Can you summarize these improvements?
Hi
We have indeed been working on improving the performance of the deep water Aquadopps.
The work goes along two lines:
a) Improving performance at low SNR. Most Doppler systems typically have a zero bias at low SNR, i.e. they tend to underestimate the velocity if the signal is too weak. Whereas all signal-processing schemes give the same result when the signal is strong, this is not the case at low SNR where the simpler schemes (for example, zero-upcrossing methods) have a significant bias.
" />
'> We have been working quite a bit on improving our own (autocovariance) scheme and we have introduced a new way of filtering. The net effect can be compared to a SNR gain of about 3dB. 
" />
'>
The firmware with the improved filtering is not in general release but is shipped with deep water systems (vers. 1.12 beta)
b) Direct improvement of the SNR
Over the last year, we have done quite a bit of work to improve the SNR.
- We have developed a new backing material that gives us about 1 dB in improved transducer efficiency
- We have developed a new head that reduces the acoustic ringing. The ringing is one of the worst issues with deep water systems because it forces the user to increase the blanking zone, thereby increasing the distance to the sampling volume. This can reduce SNR relative to a standard Aquadopps by as much as 5-8 dB. In our 2000-m Aquadopps the problem has been completely eliminated. For the 6000-m systems, we will receive the first production batch with the new transducers design in August 2002.
Historically, the systems that were shipped over the last 6 months are 3-4 dB better than those that were shipped 12 months ago. Come August, the total improvement will be at +10dB relative to the status 12 months ago. At that point we expect the Aquadopp current meter to provide accurate data in all natural bodies of water." />'>
- Atle Lohrmann
We have indeed been working on improving the performance of the deep water Aquadopps.
The work goes along two lines:
a) Improving performance at low SNR. Most Doppler systems typically have a zero bias at low SNR, i.e. they tend to underestimate the velocity if the signal is too weak. Whereas all signal-processing schemes give the same result when the signal is strong, this is not the case at low SNR where the simpler schemes (for example, zero-upcrossing methods) have a significant bias.
" />'> We have been working quite a bit on improving our own (autocovariance) scheme and we have introduced a new way of filtering. The net effect can be compared to a SNR gain of about 3dB. " />'> The firmware with the improved filtering is not in general release but is shipped with deep water systems (vers. 1.12 beta)
b) Direct improvement of the SNR
Over the last year, we have done quite a bit of work to improve the SNR.
- We have developed a new backing material that gives us about 1 dB in improved transducer efficiency
- We have developed a new head that reduces the acoustic ringing. The ringing is one of the worst issues with deep water systems because it forces the user to increase the blanking zone, thereby increasing the distance to the sampling volume. This can reduce SNR relative to a standard Aquadopps by as much as 5-8 dB. In our 2000-m Aquadopps the problem has been completely eliminated. For the 6000-m systems, we will receive the first production batch with the new transducers design in August 2002.
Historically, the systems that were shipped over the last 6 months are 3-4 dB better than those that were shipped 12 months ago. Come August, the total improvement will be at +10dB relative to the status 12 months ago. At that point we expect the Aquadopp current meter to provide accurate data in all natural bodies of water.
" />'> - Atle Lohrmann
Powered by
Ploneboard
Forgot your password?
New user?
