2026-04-01, 08:33 PM
Well this is great to hear from someone using open echo and get feedback as I had not seen much besides one youtube video.
Yes! certainly the ringdown at only 30 volts is less time, and this is the right solution, if it detects the depth is shallow, then it can use less voltage and shorter pings to reduce ringdown time, so it can work adaptively. I also intend to experiment with active dampening as well.
I think one of my main issues with open echo is it does not sweep frequency in a chirp. So correct me if I am wrong, but what I am doing is also allowing a variable bandwidth, and the pulse is sent sweeping this bandwidth. Some transducers have wider bandwidth than others. The return is then computed a FFT, along with FFT of the chirp and these are convolved in frequency space then inverted with IFFT. This mathematically produces a much sharper depth measurement, in fact it allows accuracy down to centimeters (the exact resolution depends on bandwidth) while allowing the pulse energy to spread across several milliseconds which it is not possible with a simple ping, this is the "chirp" sonar.
So this is my issue with the ti part the tuss4470, it has many limitations, and although it is much cheaper, the electronics I have are still cheaper than the transducer itself. Most transducers can handle higher voltages (up to 100 volts) which results in potentially (depending on unit again) both higher efficiency from much shorter pulse needed as well as better clarity and certainly more range, but this part limits driving to about 30 volts, probably designed for air not water??
If 30v works at 40 meters then this is less of a question, but probably at only 200khz or less though that is quite good to know. Perhaps the analog receiving part is very good. At higher frequencies like 800khz much more power might be needed and those higher frequencies have much better resolution and are desirable for mapping or imaging.
Eventually for a phased array with many elements, I dont think the TI part could ever be used as the channels would not be synchronized in time so no way to perform the beamforming computations.
If a cheaper transducer (< $20) can be found or reusing an existing one, then the open echo in absolute terms is much cheaper, especially to just get depth, and this is interesting and useful for that reason, but I am trying to produce higher resolution images with more user configurable variables, and the cost $30 or $40 more for the electronics. What do you think?
Yes! certainly the ringdown at only 30 volts is less time, and this is the right solution, if it detects the depth is shallow, then it can use less voltage and shorter pings to reduce ringdown time, so it can work adaptively. I also intend to experiment with active dampening as well.
I think one of my main issues with open echo is it does not sweep frequency in a chirp. So correct me if I am wrong, but what I am doing is also allowing a variable bandwidth, and the pulse is sent sweeping this bandwidth. Some transducers have wider bandwidth than others. The return is then computed a FFT, along with FFT of the chirp and these are convolved in frequency space then inverted with IFFT. This mathematically produces a much sharper depth measurement, in fact it allows accuracy down to centimeters (the exact resolution depends on bandwidth) while allowing the pulse energy to spread across several milliseconds which it is not possible with a simple ping, this is the "chirp" sonar.
So this is my issue with the ti part the tuss4470, it has many limitations, and although it is much cheaper, the electronics I have are still cheaper than the transducer itself. Most transducers can handle higher voltages (up to 100 volts) which results in potentially (depending on unit again) both higher efficiency from much shorter pulse needed as well as better clarity and certainly more range, but this part limits driving to about 30 volts, probably designed for air not water??
If 30v works at 40 meters then this is less of a question, but probably at only 200khz or less though that is quite good to know. Perhaps the analog receiving part is very good. At higher frequencies like 800khz much more power might be needed and those higher frequencies have much better resolution and are desirable for mapping or imaging.
Eventually for a phased array with many elements, I dont think the TI part could ever be used as the channels would not be synchronized in time so no way to perform the beamforming computations.
If a cheaper transducer (< $20) can be found or reusing an existing one, then the open echo in absolute terms is much cheaper, especially to just get depth, and this is interesting and useful for that reason, but I am trying to produce higher resolution images with more user configurable variables, and the cost $30 or $40 more for the electronics. What do you think?