Let’s start with a very brief review of single beam acoustics. A sounding starts when the echosounder electronics sends a short voltage pulse to the transducer, which converts the electrical energy to a mechanical energy in the form of an acoustic (sound) wave in the water; a ping. The transducer focuses the ping downward and almost all the energy of the ping travels within a beam.

The ping travels at the speed of sound in water. Where the sound velocity changes due to temperature or density variations, like at the boundary between velocities 1 and 2, the ping speed changes. A very small portion of the energy is reflected back upward, but the ping still travels straight down; there is no change in direction.

When the ping reaches the bottom, it encounters a large change in velocity (V3). This is because sound travels much faster in the solid bottom than it does in a liquid. A large amount of the ping energy is reflected (echoed) upward at this transition and eventually finds its way back to the transducer. The transducer converts the reflected sound back to the electrical energy. From the time delay between the outgoing and incoming pulses (and known acoustic velocity in water), depth is calculated.

Now, we take a look at multibeam sonar. In multibeam technology, a beam is sometimes called a ray, which is a mathematical term for a line with a direction. In the single beam case above, it is two directions, first down then up.

Multibeam sounding through sound velocity change; the beam is refracted upward. (V2 is greater than V1)   

With multibeam, the beams are not necessarily vertical, and that changes the situation. When a non-vertical beam encounters a change in sound velocity, not only does the ping change speed, the beam (ray) changes direction slightly. This effect is known as refraction or ray bending. When sound velocity increases (v2 > v1), the ray is bent upward. Conversely, when the sound velocity decreases (v1>v2), the ray is bent downward. Snell’s Law gives the magnitude of refraction.

V1/Sin(theta 1) = V2/Sin(theta 2) where theta 1 and 2 are the vertical ray angles in V1 and V2 respectively.

A single beam system pointed directly below the boat and a beam from a multiple transducer system angling through the water column. The Svn numbers represent the different sound velocity layers. Since the single beam is traveling perpendicular to the sound velocity layers, it will not be refracted. The (lower) red line on the multibeam shows the path of the beam without ray bending. The (farthest right) green line shows its actual path after being refracted. Notice you receive a different position and different depth, based on the effect of ray bending.

The largest source of errors in multibeam surveys is attributed to sound velocity measurements. You can typically see errors in sound velocity measurements in real time by noticing a “curling” effect on the outer beams when over flat bottoms.  

Sound velocity variations are most extreme in deep water surveys where thermal effects lead to large velocity variations. In shallow water, variations are significant in estuaries where velocity changes abruptly between fresh and salt water. Where the water is well mixed, refraction is typically not a problem.

Refraction will not introduce large errors in sounding data, even if the velocity table is not quite right, as long as the vertical ray angle doesn’t go too far beyond 45 degrees. (Under these circumstances there is less than 1- percent error, vertical and horizontal.) Beyond 45 degrees, however, the error will increase rapidly. If you have collected data in an area where the bottom is reasonably flat, but the outer beams are consistently shallower (or deeper) than the inner beams, you can be sure that refraction is being improperly compensated. The likely reason is erroneous sound velocity mea­surements.

How does this affect survey operations? Mainly, it makes the bar check procedure obsolete, which is only good for finding average sound velocity. What you will do instead, is cast a sound velocity probe to measure actual velocity variations with depth. The Veloc­ity vs. Depth information is entered into a table that is used during post-processing to compensate for refraction.

As the frequency of your EM wave increases, so does the precision of your measurement. Put hydrographers’ terms, the higher the frequency of your transducer, the more accurate your measurement will be. High-resolution side scan sonars use frequencies of 500KHZ. Multibeam systems for small launches use frequencies from 200KHZ to 450KHZ. Traditional single beam echosounder use frequencies around 200MHZ. Some hydrographers (for rea­sons to be discussed) use transducers with 24KHZ to 33KHZ. After reading this, you think everybody would be using 500KHZ transducer, but there is a price to pay for precision.

The higher the frequency of the EM wave, the greater the energy dissipation. As sound waves travel through water, their energy is dissipated by particles in the water, air bubbles, etc. High frequency sound waves quickly dissipate and cannot be used in deeper water. Deep-water transducers, used to measure depths of greater than 1000 meters, are typically in the range from 3KHZ to 12KHZ. Although not as precise as 200KHZ transducers, they can produce sound waves that can get to the bottom and back without dissipating.

The higher the frequency, the higher the reflectivity. One of the drawbacks with higher frequency (200+KHZ) transducers is they reflect off almost anything. This includes vegetation, air bubbles, fish bladders, and suspended sediments. A lower frequency transducer (e.g. 24KHZ), although slightly less precise, will allow you to pass through some of these materials to actually track the bottom. Over a soft mud, sand, silt) bottom, a low frequency transducer will generally provide deeper depths than a high frequency transducer. Over a hard (rock) bottom, the two transducers should produce almost the same depth.

The lower the frequency, the larger the transducer. The physical size of transducers has certainly been reduced over the last decades. However, this rule generally still holds true. Lower frequency transducers are heavier and larger than higher frequency transducers and sometimes can complicate the mounting procedures.

Serial interfacing can be compared to running a single pipeline. Survey information is broken into individual characters, which are then broken down into a series of ones and zeros. These ones and zeros are known as bits. Each one or zero is transmitted by chang­ing the voltage on a transmit wire. Your survey equipment may change the voltage to 5V to designate a zero, and then drop the voltage to 0V to designate a one.

Data bits and Stop bits: This series of bits is normally transmitted in series of seven or eight data bits. Each hardware device will have a setting called data bits, which defines the number of bits in each group. At the end of each group, the device inserts one or more Stop bits. This provides the equipment with a little time to pro­cess each message and prepare for the next message.

Parity: When serial transmission was first implemented, it was not as perfected as it is today. In order to check whether or not a mes­sage was correctly received, transmitting equipment would add a Parity bit. This was a single bit which would be either a zero or a one, depending on the sum of the data bits in the message group.

•If you selected Even Parity, the parity bit would be set so the sum of all of the data bits and parity bit would be an even number.

•If you selected Odd Parity, the parity bit would be set so the sum of all of the data bits and parity bit would be an odd number. This gave the receiving equipment a 50/50 chance of detecting a bad data group.

As serial equipment became more reliable, manufacturers began to eliminate the parity bit. In this case, a setting of None or No Par­ity would tell the devices not to worry about a parity bit.

Baud Rate: The final essential piece of information needed to establish communication between two devices is the Baud Rate. This is the speed, expressed in bits per second, with which the two devices send characters to each other. In order to successfully communicate, both devices need to agree as to the Baud Rate, the Data bits, Stop bits and Parity. If any of these values are not speci­fied correctly, the results may vary. For example, if you incorrectly specify the baud rate, your computer will receive what it thinks is gibberish. If you incorrectly specify the number of Stop bits, it may be able to successfully decipher 80% of the received messages.

Handshaking: The other key, essential in serial communications is called Handshaking. This is how one device tells another device that it is either ready or not ready to receive additional information. For example, most computers can send information to a plotter faster than the plotter is capable of processing it. The plotter, first, stores information in a temporary buffer until it can process it. Once the buffer becomes full, it needs some way of telling the computer to stop sending the information. This is done via Handshaking. Handshaking is normally accomplished by one of the following methods:

Xon/Xoff is preferred by some devices because it requires no additional wires, other than a transmit, receive and signal ground wires. When a device is becoming full, it sends an Xoff character (CHR$17). Upon receipt, the transmitting device stops sending information. Once the receiving device has processed enough information and can receive more information, it sends an Xon character (CHR$19). This allows the transmitting device to resume its transmission. For equipment requiring this type of handshaking, set the Flow Control to software in the COM properties dialog.

CTS/RTS (Clear to Send/Ready to Send) and DST/DTR (Data Set Ready/Data Terminal Ready) are similar methods. They each require up to two additional wires in the serial cable. The transmit­ting device uses one wire to tell the receiving device it is ready to send data. The receiving device uses the other wire to tell the transmitting device it is ready to receive data. If one, or both, of the conditions are not met, the device does not transmit. Devices that require DST/DTR handshaking are a little different. The Flow Control is still set to "hardware", but you will also need a custom cable. 

Its preferable that all handshaking be set to None. This means that as soon as a measurement is made, it is transmitted to the computer without any additional delay. Unless there are overriding reasons, all equipment, with the exception of plotters, should be set with no handshaking.

Serial Hardware

All serial ports in your computer are referenced by a location (I/O Address). Serial ports are referred to as COM ports. The first one will be called COM1:, the sec­ond one COM2:, etc. Serial ports are being phased out as standard equipment on PCs, however, they can be added by using PCMCIA or PCI serial cards. These cards come with one, two or four serial ports on a single card.

USB technology is designed to give priority to Windows® functions. This can result in extreme and incon­sistent latency issues. You cannot rely on USB to Serial connectors for time-critical data!

To configure network connections or viewing, you must know the IP Address of the broadcasting device. To find the IP Address, follow these instructions:

1.Open Command Prompt (Press the Windows key and type “cmd”, click “cmd.exe”).

2.In the Command Prompt, type “ipconfig” (without quotes) and press Enter.

3.If you are connected via WiFi, look under “Wireless LAN adapter Wi-Fi”. If you are connected via Ethernet cable, look under “Ethernet adapter Ethernet”.

Finding the IP Address of the Survey Computer

Important: If you restart/disconnect from the network, this number can and will change . If viewers are having trouble connecting, make sure they’re using the correct IP address by repeating the above steps.