ADCP Installation Guide 95A-6019-00.pdf

Ocean Surveyor Ocean Observer

Installation Guide

P/N 95A-6019-00 (December 2002)

RD Instruments Acoustic Doppler Solutions

Table of Contents 1 Introduction 2 Ocean Surveyor/Observer System Components 3 Transducer Mounting Considerations

3.1 Location

3.2 Orientation

3.3 Floating Objects

3.3.1 Acoustic Window

3.3.2 Sea Chest

3.3.3 Fairing

3.4 Air Bubbles

3.5 Flow Noise

3.6 Corrosion and Cathodic Disbondment

3.7 Ringing

3.8 Acoustic Isolation

3.9 Maintenance

4 Using Acoustic Windows

4.1 Background - Should I use an Acoustic Window?

4.1.1 Advantages

4.1.2 Disadvantages

4.2 What Window should I use?

4.2.1 Are there any Other Windows that I can consider?

4.3 What are the Acoustic Window Installation Considerations?

4.3.1 What should the Window Shape be?

4.3.2 Can I Add Strengthening to the Window?

4.3.3 How do I Secure the Window to the Well Opening?

4.3.4 How to Prevent the Window From Cracking When Going Into Dry Dock?

4.3.5 How Much Space should I Have Between the Window and Transducer?

4.4 What Other Issues should be Considered When Using an Acoustic Window?

4.4.1 What Fluid Should I Fill the Sea Chest With?

4.4.2 How Much Fluid Should I Use in the Sea Chest?

4.4.3 Should I Use Absorption Material When Using a Window?

4.4.4 Do I Need to Worry About Corrosion When Using a Window?

4.4.5 Do I Need to Worry About Air Bubbles When Using a Window?

4.4.6 Do windows Improve Flow Noise Problems?

4.5 What Maintenance is required when using Windows?

4.5.1 How Often Should I Inspect the Window?

4.5.2 If I find that the Window is Damaged Can I Keep Operating the ADCP?

4.5.3 Does the Use of a Window Effect My Warranty?

4.6 Insertion Loss

5 Conceptual Transducer Well Design 6 Electronics Chassis Mounting Considerations

6.1 Cabling Considerations

6.2 I/O Cable

6.3 I/O Cable Identification

6.4 Dry Connector Installation

6.4.1 Tools and Equipment Requirements (RDI Supplied)

6.4.2 Tools and Equipment Requirements (User Supplied)

6.4.3 Installation Instructions

6.5 Using Trigger-In and Trigger-Out

7 Computer Considerations 8 Gyrocompass Interface Considerations

8.1 Overview of the ADCP Gyro Interface

8.1.1 Determining the Synchro Stator Voltage

8.1.2 Determining the Size of the Scaling-Resistor Pack

8.1.3 Installing the Scaling-Resistor Pack and W-Jumpers

8.1.4 Determining and Setting the Synchro Turns Ratio

8.2 Stepper Interface

8.3 Testing the Gyro Interface

8.4 Gyro Cable

9 Navigation Interface Considerations 10 Power Considerations 11 Installation Procedures (Overview) 12 Alignment Procedures (Overview) 13 Specifications

13.1 Water Velocity Specifications

13.2 Profile Parameters

13.3 Bottom Track Specifications

13.4 Echo Intensity Profile

13.5 Transducer and Hardware Specifications

13.6 Internal Sensors

13.7 System Power Specifications

13.8 Environmental Specifications

14 Outline Installation Drawings Ocean Surveyor/Observer Electronics Chassis - 96A-6000 Ocean Surveyor/Observer 75kHz - 96A-6007 Ocean Surveyor/Observer 75kHz, Wide Flange - 96A-6011 Ocean Surveyor/Observer 150kHz - 96A-6008 Ocean Surveyor/Observer 38kHz – 96A-6009 Ocean Surveyor/Observer 150kHz, Flanged – 96A-6010

List of Figures Figure 1. Conceptual Drawing of an Acoustic Window with Mounting Ring Figure 2. Location of the Window Bolt Hole Circle Diameter Figure 3. Draining holes locations around beams 1-4 intersection with the window Figure 4. 38kHz Insertion Loss (one-way) with a 19.1mm window at 0°C Figure 5. 38kHz Insertion Loss (one-way) with a 38.1mm window at 0°C Figure 6. 75kHz Insertion Loss (one-way) with a 9.5mm window at 0°C Figure 7. 75kHz Insertion Loss (one-way) with a 9.5mm window at 0°C Figure 8. 75kHz Insertion Loss (one-way) with a 19.1mm window at 0°C Figure 9. 75kHz Insertion Loss (one-way) with a 19.1mm window at 0°C Figure 10. 150kHz Insertion Loss (one-way) with a 9.5mm window at 0°C Figure 11. 150kHz Insertion Loss (one-way) with a 9.5mm window at 0°C Figure 12. 150kHz Insertion Loss (one-way) with a 19.1mm window at 0°C Figure 13. 150kHz Insertion Loss (one-way) with a 19.1mm window at 0°C Figure 16. Inside Vessel Mounting - OS 38khz Transducer Figure 17. Underneath Vessel Mounting - OS 38khz Transducer Figure 18. Inside Vessel Mounting - OS 75khz Transducer Figure 19. Underneath Vessel Mounting - OS 75khz Transducer Figure 20. Inside Vessel Mounting - OS 150khz Transducer Figure 21. Underneath Vessel Mounting - OS 150khz Transducer Figure 24. I/O Cable Figure 25. Typical Ocean Surveyor/Observer Interface Cable Layout (Overview) Figure 26. Typical Ocean Surveyor/Observer Interface Cable Layout (Detail View) Figure 27. I/O Cable Wiring (73A-6010) Figure 28. OS2 Versus an OS1 I/O Cable Wet-End Connector Figure 29. Gyro Interface Board

Figure 30. Cable, Gyro Synchro/Stepper to Electronic Chassis (J2) Figure 31. Transducer Misalignment Reference Points

List of Tables Table 1: Expected Loss for ADCPs with 30Degree Beam Angle Table 2: Expected Loss for ADCPs with 20Degree Beam Angle Table 3: Wavelength of sound in seawater (1500 m/s sound speed) Table 4: Trigger-In Input/Output Signals Table 5: Minimum Computer Hardware Requirements (two COM Ports only) Table 6: Recommended Computer Configuration Table 7: Gyro Interface Options by Model Table 8: Acceptable Gyro Interface Configurations Table 9: Standard Synchro Stator Voltages and Scaling Resistance Table 10: Natel Chip, Resistor Pack, and W-jumper Sockets Table 11: Gyro Interface Switch 1 Settings Table 12: Gyro Interface Switch 2 Settings Table 13: Stepper Voltage Range Table 14: Gyro Interface Troubleshooting Guide Table 15: Water Profiling – Long Range Mode Table 16: Water Profiling – High Precision Mode Table 17: Water Profile Maximum Ping Rate Table 18: Nominal Bottom Track Altitude Table 19: Outline Installation Drawings

Ocean Surveyor/Observer Installation Guide

P/N 95A-6019-00 (December 2002) page 1

Acoustic Doppler Solutions

1 Introduction This book is a guide for installing a RD Instruments (RDI) Ocean Sur-veyor/Observer Acoustic Doppler Current Profiler aboard a vessel (ship), or platform. Use this section to plan your installation layout. You also can use this book to see what requirements you must consider before purchasing an ADCP. We recommend you distribute this book to your organization’s de-cision-makers and installation engineers.

We are not experts in installing the ADCP aboard a ship. There are too many installation methods. We suggest you seek expert advice in this area because of its importance in ADCP performance. However, we can give you information about how others have installed their systems. In return, we do appreciate receiving information about your installation and the results.

2 Ocean Surveyor/Observer System Components The standard Ocean Surveyor/Observer is an AC-powered, two-part unit. It consists of a bronze transducer assembly mounted in the ship’s hull and an electronics chassis you can place in the ship’s lab. Several options are available.

The basic system includes:

• Transducer assembly - Contains the transducer ceramics and electronics. Standard acoustic frequencies are 38, 75, and 150kHz. See the outline drawings for dimensions and weights.

• Electronics chassis - Contains the processing electronics.

page 2 RD Instruments

• Maintenance kit - Contains spare parts and basic tools needed for routine maintenance.

• Ocean Surveyor/Observer technical manual and user’s guide.

• Waterproof dummy plug.

• VmDas Software - Controls the ADCP and displays its data through a Windows compatible computer.

• Utilities software - Several programs to help you talk directly to the ADCP (BBTalk), view data (WinADCP) and other DOS util-ity programs (RDI Tools).

Standard options include:

• Input/Output (I/O) cables and connectors (length determined by customer).

• Gyrocompass (gyro) interface - Connects the ship’s gyro to the electronics chassis.

3 Transducer Mounting Considerations You must consider several potential problems before installing the trans-ducer assembly. Read this section before deciding where to install the transducer assembly. See the outline installation drawings (see “Outline Installation Drawings,” page 61) for specifications on our standard Ocean Surveyor/Observer transducer.

3.1 Location

Ideally, you want to install the transducer assembly:

• Where it is accessible both internally (for access to transducer electronics) and externally (to remove biofouling).

• Where the I/O cable length is 100 m (328 feet) or less.

• Away from shipboard protrusions that reflect ADCP energy. Al-low for a reflection-free clearance of 15° around each beam (see the outline installation drawings).

• Away from other acoustic/sonar devices, especially those operat-ing at the same frequency (or harmonic) of the ADCP.

• Close to the ship’s fore-to-aft centerline. As distance from the centerline increases, vertical accelerations caused by the roll of the ship also increase. These accelerations can cause additional uncertainties in ADCP velocity measurements.

P/N 95A-6019-00 (December 2002) page 3

3.2 Orientation

The 38kHz oval transducer should always be mounted with the transducer’s beam 3 mark on axis (forward). If you have a round Ocean Sur-veyor/Observer transducer, then we recommend you mount the transducer head with the Beam 3 mark rotated to a ship-relative angle of 45°.

Figure 27, page 56 shows the beam orientation. This causes the magnitude of the signal in each beam to be about the same. This improves error rejec-tion, reduces the effect of ringing (see “Acoustic Isolation,” page 6), and increases the ADCP’s effective velocity range by a factor of 1.4. If you align Beam 3 at an angle other than zero, you must nullify this offset (see “Alignment Procedures (Overview),” page 55). You can do this through our VmDas program or using a direct command.

Use the ship’s roll and pitch reference to mount the transducer head as level as possible. If the head is not level, depth cell (bin) mapping will be incor-rect. Large misalignments can cause large velocity measurement errors. If you cannot mechanically make the transducer head level, you can use VmDas to enter offset values for roll and pitch, or use direct commands for roll/pitch offsets (see “Alignment Procedures (Overview),” page 55).

3.3 Floating Objects

Our transducer assembly is sturdy, but we did not design it to withstand col-lisions with all floating objects. We strongly suggest you use one of the following.

3.3.1 Acoustic Window

While we do not fully understand windows, we do believe that windows can be used to produce overall performance improvements in vessel-mounted ADCPs. Additionally, if the ship operates where there is danger of barnacle damage or a high density of ice or other floating objects, then the use of an acoustic window is the only option. For more information, see “Using Acoustic Windows,” page 8.

3.3.2 Sea Chest

A sea chest is a fixture that surrounds and holds the transducer head, pro-tecting it from debris in the water. The bottom of the sea chest must be open to seawater to allow the acoustic beams to pass through freely. For more information, see “Conceptual Transducer Well Design,” page 26.

3.3.3 Fairing

A fairing is a structure that produces a smooth outline and reduces drag or water resistance. The fairing also diverts floating objects away from the page 4 RD Instruments transducers. A fairing shaped like a teardrop, slopped such that the leading edge (closer to the bow) is higher than the back edge, and extends below the hull (typically 12 inches) will divert the air bubbles away from the trans-ducer faces.

3.4 Air Bubbles

Design your installation to minimize the volume of air bubbles in the path of the acoustic beams. Air bubbles attenuate (weaken) the signal strength and reduce the ADCP profiling range. Ships with a deep draft or a non-flat bottom have fewer problems with bubbles. Ways to reduce bubble flow vary with ship characteristics, but two options are available. Mount the transducers below or away from the bubble layer.

• The flow layer is usually within the first two feet below the hull.

Bubbles can be trapped in this layer. Mounting the transducer head amidship on the fore-to-aft centerline may help. For ships with propulsion systems that make large amounts of bubbles, use a mounting technique that lets you lower the transducer head be-low the hull while underway.

NOTE. If you use locally made or existing extension hardware instead of the hardware available from RDI, you may need to make an adapter plate to connect your hardware to our transducer head. See “Outline Installation Drawings,” page 61 for the bolt hole locations.

• Divert the bubble layer so it flows around the transducers - You can use fairings to alter the bubble flow (see “Fairing,” page 3).

An acoustic window (see “Acoustic Window,” page 3) may help reduce the bubble problem, but can cause ringing (see “Acoustic Isolation,” page 6) and attenuation problems.

3.5 Flow Noise

Water flowing over the transducer faces increases the acoustic noise level, which decreases the profiling range of the ADCP. You can reduce the flow across the transducer faces with a sea chest, fairing, or acoustic window.

3.6 Corrosion and Cathodic Disbondment

Never attach anodes directly to the transducer head. Additional anodes or impressed voltage systems can cause the urethane to separate from the transducer (cathodic disbondment) or cause the material of the transducer to break down. Standard anode protection used for the ship should be in-stalled outside of the well of the transducer head. Mounting of ship’s stan-dard anode protection outside of the transducer well will typically not cause any problems. Our transducers are made of a material that has shown to

P/N 95A-6019-00 (December 2002) page 5 corrode very little over time when the above precautions are met. The cor-rosion can be further reduced if the well is covered with a window and then filled with fresh water.

3.7 Ringing

The ADCP transmits an acoustic pulse into the water. The main lobe of this pulse bounces off particles in the water and the signals returned from these particles are used to calculate the velocity of the water.

As stated, the main lobe of the transmitted pulse is what we are using to process and calculate a velocity. The transmitted pulse, however, is made up of many side lobes off the main lobe. These side lobes will come in con-tact with metal of the transducer beam itself and other items in either the water or the well.

The energy from the side lobes will excite the metal of the transducer and anything bolted to the transducer. This causes the transducer and anything attached to it to resonate at the system’s transmit frequency. We refer to this as “ringing.”

If the ADCP is in its receive mode while the transducer is ringing then it will receive both the return signals from the water and the “ringing.” Both of these signals are then processed by the ADCP. The ringing causes bias to the velocity data.

All ADCPs “ring” for some amount of time. Therefore, each ADCP re-quires a blanking period (time of no data processing) to keep from process-ing the ringing energy. Each ADCP frequency has a different typical ring-ing duration. The typical ringing period for each ADCP frequency is as fol-lows; 38kHz is 16 meters, 75kHz is 8 meters, and 150kHz ADCPs is 4 me-ters. These typical ringing values are recommended as the minimum setting for all ADCPs using default setups.

It should be noted, on some installations the effects of ringing will last longer than the recommended settings above. For example, the effects of ringing will last longer if the transmit signal becomes trapped inside the transducer well. This can occur because the well itself is ringing with the transducer or when windows covering the opening of the well reflect the signal back inside the well.

The window causes the transmit signal to reflect back into the well due to the difference in impedance between the window and the water. When the transmit signal is reflected in the well it becomes trapped and this results in longer ringing periods. To keep from processing this signal, the blanking period must be increased. Lining the inside walls of the well with a sound absorbing material aid in dampening the ringing effect.

page 6 RD Instruments

3.8 Acoustic Isolation

Try to minimize the acoustic coupling between the transducer head and the ship. Without adequate acoustic isolation, the transducer output will “ring” throughout the ship and feeds back into the ADCP receive circuits. Ringing causes bias errors in water-track velocities and results in the loss of data in the closest depth cells (bins). Reflections inside a sea chest with an acous-tic window also can cause ringing.

You can attain acoustic isolation several ways. At a minimum, use gaskets to isolate all contact points between the ship and the transducer head. De-sign your installation for:

• A minimum number of contact points between the transducer head and the ship.

• Minimal contact area.

• Single points of contact for positioning and support (when possible).

You also should try to separate the transducer head from the ship using in-termediate connections. This is because direct connections transfer the most acoustic energy. Texas A & M used the following installation tech-nique and had minimal ringing problems.

• Transducer head mounted to a thin steel plate

• Steel plate positioned with three pins set into mounting holes on the hull; pins isolated with gaskets

• Steel plate held in place with four I-beams welded to a frame

• Frame bolted to another frame and separated by gaskets

• Second frame bolted to the ship and separated by gaskets

Acoustic isolation from other acoustic devices on the ship is also necessary.

You can do this using the following techniques.

• Mount the other acoustic devices as far apart as possible.

• Make sure neither the main lobes nor the side lobes of the acoustic devices point at the transducers, including acoustic reflections.

• Do not to operate devices that use the same frequency or a har-monic of the ADCP’s frequency.

3.9 Maintenance

The Maintenance guide explains routine maintenance procedures. You rarely need access to the electronics inside the transducer head. However, P/N 95A-6019-00 (December 2002) page 7 one external maintenance item is important enough to mention here as it may affect how you install the transducer head.

Objects deployed within about 100 meters (328 feet) of the surface are sub-ject to the buildup of organic sea life (biofouling). This means Ocean Sur-veyor/Observers are subject to biofouling. Soft-bodied organisms usually cause no problems, but hard barnacle shells can cut through the urethane transducer face causing transducer failure and leakage into the ADCP.

The best-known way to control biofouling is cleaning the ADCP transducer faces often. However, in many cases this is not possible. The other alterna-tives include the use of a window or some sort of anti-foulant protection.

Some of our users have had success applying a thin coat (≈4 mm; ≈0.16 in.)

of either a 50:50 mix of chili powder and Vaseline or chili powder and sili-cone grease to the transducer faces. The chili powder should be the hottest that can be found. Water flowing across the transducers will wash this mix away over time. The silicone mixture tends to last longer.

Some organizations may decide to use antifouling grease. However, most antifouling greases are toxic and may cause problems. Recent tests suggest antifouling grease may cause the urethane on the transducer faces to de-velop cracks. Warmer temperatures accelerate this effect.

The other method is to use antifoulant paint. At present, we recommend the following antifouling paint manufacturer and paint brand: Courtalds Fin-ishes Interlux brand paints, US Telephone: 908-686-1300, Web Page:

www.interlux.com. Contact the antifouling paint manufacturer for prepara-tion and application procedures for this and other antifoulant paints.

CAUTION.

1. Read the Material Safety Date Sheet before using any of the listed solvents and paints.

2. Some antifouling coatings may not be legal for use in all areas. Check with your local environmental agency before using the antifouling paint.

3. Do not arbitrarily use antifouling paints. Be aware that antifouling paints can accelerate the dezincification corrosion of brass. Once initiated, dezincificatioin will rapidly destroy the transducer.

4. RDI no longer recommends the use of Nopcocide for the prevention of biofouling. If using antifouling grease, remove it immediately after recovering the ADCP.

5. Antifouling grease is toxic. Read the product safety data sheet before using the grease.

Wear gloves and a face shield when applying the grease. If the skin comes in contact with the grease, immediately wash the affected area with warm, soapy water.

6. When possible, do not coat the transducer faces with cuprous oxide or related paints that contain chemicals such as copper, chrome, or arsenic. These paints advance the corrosion of the transducer assembly and will cause the urethane to separate from the transducer cups.

7. All US Coastal States prohibit the use of tributyl-tins on boat hulls. The European Economic Commission has released a draft directive that would prohibit the use of many organo-tins after July 1989. We strongly recommend you obey your local laws.

http://www.interlux.com/ page 8 RD Instruments

4 Using Acoustic Windows RD Instruments (RDI) builds Acoustic Doppler Current Profilers (ADCP) for use on vessels. Installation of an ADCP in a vessel is done in many ways, but typically the ADCP transducer is mounted inside of a sea chest or well. The opening of the sea chest or well can be open to the ocean or it can be covered by an acoustic window.

An acoustic window is a covering that can seal the opening of the sea chest but still allow the acoustic signal (both transmit and received signals) to be transferred through the window. The type, thickness, orientation, and other installation issues of the acoustic window are important to understand. If the wrong material is used or the wrong installation used then the perform-ance obtained by the ADCP will be severely limited.

4.1 Background - Should I use an Acoustic Window?

Like any vessel-mount, acoustic system, the performance of the ADCP is sensitive to acoustic noise. For best performance, the transducer is mounted in its own well, recessed in the vessel hull, with an opening slightly larger than the transducer. An Acoustic Window, mounted across the well open-ing, is required to isolate the transducer face from flow noise, as the vessel moves through the water. Acoustic windows (or simply windows) can pro-duce overall performance improvements in vessel-mounted ADCPs through the following advantages.

4.1.1 Advantages

• Well will not fill with air bubbles caused by the ship moving through the surface water.

• Flow noise is reduced.

• The well can be filled with fresh water to limit corrosion.

• Barnacles cannot grow on the transducer faces. Barnacle growth is the number one cause of failure of the transducer beams.

• The transducer is protected from debris floating in the water.

Although these advantages are important, it should be known that if the wrong window is used or if the window is not installed properly then the following disadvantages are possible.

4.1.2 Disadvantages

• The range of the ADCP can be reduced because the window will absorb some of the transmit and receive energy.

P/N 95A-6019-00 (December 2002) page 9

• The transmit signal could be reflected into the well, causing the well to “ring” like a bell. This will cause the data being collected during the ringing to be biased. Some ships have reported a loss in range as great as 50 meters. Applying sound absorbing mate-rial on the well walls may dampen the ringing.

• The transmit signal could be reflected off of the window and back into the other beams.

However, even though there are disadvantages possible our experience has shown that when the correct window is used and it is properly installed that the window advantages are far more important. The remainder of this Ap-plication Note will focus on how to choose the window for your vessel, how to mount the window, how to maintain the window, and any other associ-ated concerns when using a window.

4.2 What Window should I use?

While we cannot claim to understand every window, we do believe that we can recommend a material that will work. We have developed a simple model for an acoustic window made from polycarbonate material. Over the past 2 decades we have obtained feed back from customers that has allowed us to prove the model is a fair estimation of what to expect for performance.

Polycarbonate was chosen because it can provide enough strength for most installations, is readily available in most countries, it has been shown to last a long time (over 7 years in some installations), and it can be used on all ADCP models (NarrowBand (NB), BroadBand (BB), WorkHorse (WH), and Ocean Surveyor (OS)).

The type of ADCP model is very important when choosing a window. The bandwidth of the acoustic signal from the ADCP must be maintained. Dif-ferent window materials have different losses over a band of frequencies.

As an example, the Ocean Surveyor/Observer ADCP uses a bandwidth of 6% or 1% about the system’s center frequency. A BroadBand or WorkHorse ADCP uses a bandwidth of 25% or 6% about the system’s center frequency.

The material polycarbonate has a fairly uniform loss about these frequency bandwidths.

It should be noted that we have no knowledge about the variability of poly-carbonates. And so, the acoustic model that we run is for a particular poly-carbonate manufactured by Zelux. This is a window-grade, polycarbonate and has a high tensile strength (~9000psi) to resist cracking.

Even when choosing this particular window it is important to choose the proper thickness of window material. A window will absorb sound and re-duce the range of the ADCP. Therefore, we always recommend using the thinnest window possible. However, depending on your application a thicker material may be necessary. The following table indicates the ex-page 10 RD Instruments pected loss (two-way) of polycarbonate at different frequencies and thick-ness.

Table 1: Expected Loss for ADCPs with 30Degree Beam Angle Frequency

(kHz) Thickness mm (in.)

One-way loss @ 0°,20°,40°C (dB)

Two-way loss @ 0°,20°,40°C (dB)

Expected Loss in Range (meters)

38 76.2 (~3.0) 2.7 2.6 2.3 5.4 5.2 4.6 173

38 63.5 (~2.5) 3.0 2.9 2.5 6.0 5.8 5.0 192

38 50.8 (~2.0) 2.9 3.2 2.9 5.8 6.4 5.8 205

38 38.1 (~1.5) 1.4 1.2 1.0 2.8 2.4 2.0 90 38 25.4 (~1.0) 2.9 3.3 3.3 5.8 6.6 6.6 211

38 19.1 (~0.75) 1.0 0.9 0.8 2.0 1.8 1.6 64 38 12.7 (~0.5) 5.3 5.8 5.6 10.6 11.6 10.2 371

38 9.5 (~0.375) 1.8 1.8 1.8 3.6 3.6 3.6 115

38 6.4 (~0.25) 0.7 0.7 0.6 1.4 1.4 1.2 45

75 76.2 (~3.0) 4.2 4.3 3.8 8.4 8.6 7.6 138

75 63.5 (~2.5) 3.9 4.0 3.5 7.8 8.0 7.0 128

75 50.8 (~2.0) 3.6 3.6 3.0 7.2 7.2 6.0 115

75 38.1 (~1.5) 2.7 2.6 2.3 5.4 5.2 4.6 83

75 25.4 (~1.0) 3.1 3.3 2.9 6.2 6.6 5.8 106

75 19.1 (~0.75) 1.4 1.2 1.0 2.8 2.4 2.0 45

75 12.7 (~0.5) 3.1 3.5 3.3 6.2 7.0 6.6 112

75 9.5 (~0.375) 1.0 0.8 0.7 2.0 1.6 1.4 32 75 6.4 (~0.25) 5.9 6.3 5.5 11.8 12.6 11.0 202

150 50.8 (~2.0) 5.0 5.2 4.6 10.0 10.4 9.2 83

150 38.1 (~1.5) 4.2 4.4 3.8 8.4 8.8 7.6 70

150 25.4 (~1.0) 3.6 3.6 3.0 7.2 7.2 6.0 58

150 19.1 (~0.75) 2.7 2.6 2.3 5.4 5.2 4.6 43 150 12.7 (~0.5) 3.1 3.3 2.9 6.2 6.6 5.8 53

150 9.5 (~0.375) 1.4 1.2 1.0 2.8 2.4 2.0 22 150 6.4 (~0.25) 3.2 3.6 3.3 6.4 7.2 6.6 58

300 25.4 (~1.0) 5.0 5.2 4.5 10.0 10.4 9.0 42

300 19.1 (~0.75) 4.2 4.3 3.8 8.4 8.6 7.6 34

300 12.7 (~0.5) 3.6 3.6 3.0 7.2 7.2 6.0 29

300 9.5 (~0.375) 2.7 2.6 2.3 5.4 5.2 4.6 22 300 6.4 (~0.25) 2.9 3.4 3.2 5.8 6.8 6.4 27

RDI’s recommended thickness is in blue bold. RDI’s recommended maxi-mum thickness is in red italic and bold in the above table. All other items will result in poor overall performance or a loss in range that most custom-ers find unreasonable.

One-way insertion loss curves for all items above in bold (RDI’s recom-mended thickness) are found in Appendix A of this application note. All other plots are available from RDI upon request.

P/N 95A-6019-00 (December 2002) page 11

Note all of the losses and expected ranges are estimated and some of the assumptions we make may not be true in your installation. However, based on several actual installations the values shown have proven to be good es-timations. Your actual loss may be higher or lower than what is shown.

Table 2: Expected Loss for ADCPs with 20Degree Beam Angle Frequency

(kHz) Thickness mm (in.)

One-way loss @ 0°,20°,40°C (dB)

Two-way loss @ 0°,20°,40°C (dB)

Expected Loss in Range (meters)

38 76.2 (~3.0) N/A N/A N/A N/A N/A N/A N/A

38 63.5 (~2.5) N/A N/A N/A N/A N/A N/A N/A

38 50.8 (~2.0) N/A N/A N/A N/A N/A N/A N/A

38 38.1 (~1.5) N/A N/A N/A N/A N/A N/A N/A 38 25.4 (~1.0) N/A N/A N/A N/A N/A N/A N/A

38 19.1 (~0.75) N/A N/A N/A N/A N/A N/A N/A 38 12.7 (~0.5) N/A N/A N/A N/A N/A N/A N/A

38 9.5 (~0.375) N/A N/A N/A N/A N/A N/A N/A

38 6.4 (~0.25) N/A N/A N/A N/A N/A N/A N/A

75 76.2 (~3.0) 3.1 3.2 2.8 6.2 6.4 5.6 102

75 63.5 (~2.5) 2.7 2.8 2.6 5.4 5.6 5.2 90

75 50.8 (~2.0) 2.6 2.6 2.3 5.2 5.2 5.6 90

75 38.1 (~1.5) 2.0 2.0 1.8 4.0 4.0 3.6 64

75 25.4 (~1.0) 2.2 2.1 1.8 4.4 4.2 3.6 70

75 19.1 (~0.75) 0.9 0.9 1.0 1.8 1.8 2.0 32

75 12.7 (~0.5) 2.8 2.9 2.4 5.6 5.8 5.4 93

75 9.5 (~0.375) 1.0 0.9 0.8 2.0 1.8 1.6 32 75 6.4 (~0.25) 4.2 3.7 2.7 8.4 7.4 5.4 134

150 50.8 (~2.0) 3.6 3.8 3.4 7.2 7.6 6.8 61

150 38.1 (~1.5) 3.1 3.2 2.8 6.2 6.4 5.6 51

150 25.4 (~1.0) 2.6 2.6 2.3 5.2 5.2 4.6 42

150 19.1 (~0.75) 2.0 2.0 1.8 4.0 4.0 3.6 32 150 12.7 (~0.5) 2.2 2.1 1.8 4.4 4.2 3.6 35

150 9.5 (~0.375) 0.9 0.9 0.9 1.8 1.8 1.8 14 150 6.4 (~0.25) 2.9 2.9 2.3 5.8 5.8 5.6 46

300 25.4 (~1.0) 3.6 3.8 3.4 7.2 7.6 6.8 30

300 19.1 (~0.75) 3.1 3.2 2.8 6.2 6.4 5.6 26

300 12.7 (~0.5) 2.6 2.6 2.3 5.2 5.2 4.6 21

300 9.5 (~0.375) 2.0 2.0 1.8 4.0 4.0 3.6 16 300 6.4 (~0.25) 2.2 2.1 1.8 4.4 4.2 3.6 18

RDI’s recommended thickness is in blue bold. RDI’s recommended maxi-mum thickness is in red italic and bold in the above table. All other items will result in poor overall performance or a loss in range that most custom-ers find unreasonable.

page 12 RD Instruments

One-way insertion loss curves for all items above in bold (RDI’s recom-mended thickness) are found in “Insertion Loss,” page 20. All other plots are available from RDI upon request.

Note all of the losses and expected ranges are estimated and some of the assumptions we make may not be true in your installation. However, based on several actual installations the values shown have proven to be good es-timations. Your actual loss may be higher or lower than what is shown.

4.2.1 Are there any Other Windows that I can consider?

RDI has only limited experience with other materials. As a result there is not much information we can provide about other materials. However, we can state that different materials will behave differently, depending on both the frequency and bandwidth of the acoustic ADCP signal. The absorption curves of various materials have significant amplitude fluctuations with fre-quency, which can change in both frequency and amplitude with changes in temperature.

Important acoustic properties of the window include acoustic refractive in-dex (which should be as close as possible to that of water), insertion loss (which should be as small as possible) and speed of sound. There are two acoustic refractive indices: one for shear waves and one for plane waves.

The acoustic refractive indices are simply the ratios of speed of sound in water to speed of sounds in the material. Insertion loss combines absorption and reflection of sound, and it depends on both the thickness and the mate-rial properties of the window. In particular, you should avoid using window thickness equal to odd multiples of shear mode quarter-waves (Dubbelday and Rittenmeyer, 1987; Dubbleday, 1986). Refer to Selfridge (1985) and Thompson (1990) for more information. Note that the speeds of sound in plastics decrease with increasing temperature and that causes the resonant frequencies to shift. This can be a large effect. Neither Selfridge nor Thompson has much information on the temperature coefficients of sound speeds.

• The life of the material must also be considered as well as its overall strength. We have had customers design their own win-dows out of Kevlar. They required Kevlar because they required a material that was very strong both for temperature and for strength against heavy seas, objects in the water, and striking the bottom or ice. Kevlar can provide this strength without having to be very thick thus minimizing loss.

• Kevlar windows have been successfully built and used by 2 dif-ferent institutes (Monterey Bay Aquarium Research Institute (MBARI) in the United States, and the Institut National des Sci-ences de l'Univers (INSU) in France. The procedure to build the

P/N 95A-6019-00 (December 2002) page 13 window is not known by RDI. The properties of the Kevlar windows are not well understood and so a lot of experimentation with different thickness windows was required before these cus-tomers were satisfied with the Kevlar window.

• INSU used a graduate student, Roche Frederic in 1997, to per-form a study to determine the best thickness and composite of Kevlar to provide a window for a BB VM150kHz ADCP. The report states that the material KEVLAR K49 made with Resine Vinylester ATLAC 580 was used. The following French com-pany produced this material:

Brest-Composite Industrie 124 Rue Auatole Frauce 29200 – Brest, France Tel: 02-98-05-19-09 Fax: 02-98-34-06-02

RDI only knows that this single window was produced for INSU and does not know of any others who are using this material. It is RDI’s understand-ing that the above-mentioned company can produce the Kevlar window but cannot give the acoustic properties required to determine the losses through the window. Contacting this company is done with the knowledge that RD Instruments is not recommending them, but only offering this as a source for the material.

4.3 What are the Acoustic Window Installation

Considerations?

In Section 4.2 we provided the recommended window material and thick-ness. In this section we will provide installation recommendations. Instal-lation of the window must be done properly so that the best performance is possible. The following discussion is broken into sections so that each point can be considered individually.

4.3.1 What should the Window Shape be?

The window should be smooth without cracks or deformities. Typically the window is round and of a diameter that is large enough to clear all four beams. To determine the proper diameter of the window see “Conceptual Transducer Well Design,” page 26.

The acoustic window should be flat and parallel to the transducer mounting plate. This will result in a constant angle of 20 or 30 degrees (depending on the transducer beam angle) to the transducer on both the inside and outside window face.

page 14 RD Instruments

Dome shaped windows have been used successfully. However, if the water temperatures inside the window and outside the window are not the same, all four beams will be refracted and actual velocity components will be ro-tated into a new coordinate system. In particular, some of the horizontal ve-locity will appear as a vertical velocity.

4.3.2 Can I Add Strengthening to the Window?

Adding a strengthening member across the window is not recommended because this can cause similar behavior as a cracked window (see “If I find that the Window is Damaged Can I Keep Operating the ADCP?,” page 19) or can actually block the acoustic transmit and receive signals.

4.3.3 How do I Secure the Window to the Well Opening?

We recommend that a steel ring around the outside of the window be used because you do not want the screw heads to come in direct contact with the window material as it may crack under the strain.

It is recommended that window be designed so that the ring will sit flush with the entire window face as shown in the Figure 1. Flat headed bolts or recessed bolts should be used. All of these will maintain a smooth surface around the entire window and will prevent any chance for cavitation (see “Do I Need to Worry About Air Bubbles When Using a Window? ,” page 18 for more information).

Polycarbonate Window

Steel Ring

Figure 1. Conceptual Drawing of an Acoustic Window with Mount-ing Ring

Do not thread the polycarbonate window. Use bolt through holes spaced evenly around the window. The number of bolt through holes (typically 16 to 24 holes) should be enough to prevent leakage and will provide equal pressure on the window to prevent cracking.

The bolt circle should be located a distance from the edge of the window that is a minimum of twice the diameter of the bolt through holes, see Figure 2, page 15.

P/N 95A-6019-00 (December 2002) page 15

Minimum twice the diameter of the bolt through hole.

Figure 2. Location of the Window Bolt Hole Circle Diameter

4.3.4 How to Prevent the Window From Cracking When Going

Into Dry Dock?

Using as thin a window as possible may mean that the window will not be able to support the water inside the sea chest when the vessel goes into dry dock. This means that you must either be sure to have a way to drain the sea chest prior to going into dry dock or allow a way for the water to drain out of the sea chest during dry dock. The former must be done as part of the sea chest design and the latter can be accomplished by placing holes in the window face.

The holes in the window face will allow water to freely flow in and out of the well. However, drilling holes in the window will increase your chances of flow noise, air bubbles in sea chest, corrosion, bio-fouling, and will make the sea chest non-hydrostatic. The bio-fouling will require that you have regular transducer inspections and cleanings. If you make the sea chest non-hydrostatic then in heavy seas the window can crack as it flexes from wave slamming. Dave Taylor Model Basin has measured slamming pres-sures as high as 300 psi with durations of a few milliseconds. If the sea chest behind the window is hydrostatic, no pressure gradient will exist across the window and no substantial deflections will occur.

With those considerations in mind you may still want to drill holes in the window. If you drill holes in the window then you must make sure that they are outside of the acoustic beam pattern of the beams, see Figure 3. As long as the holes remain outside of the beams they can be as large as is required to prevent the window from flexing when in heavy seas.

page 16 RD Instruments

Locations for draining holes.

4 3

Figure 3. Draining holes locations around beams 1-4 intersection with the window.

4.3.5 How Much Space should I Have Between the Window

and Transducer?

Never allow the transducer to touch the window. Separation is good for re-ducing the strength of the multiple fields of flow noise. However, we must limit the separation to prevent the reflection of a beam off of the window into another beam. This causes cross talk between the beams.

Therefore, with all ADCP models and frequencies the recommended dis-tance between the transducer and the inside face of the window should be between 6.4mm to 12.7mm. This will ensure that there is no cross talk be-tween beams and will provide adequate spacing to reduce flow noise. See “Conceptual Transducer Well Design,” page 26 for more sea chest design issues.

4.4 What Other Issues should be Considered When

Using an Acoustic Window?

Once you decide to use a window there are many issues that no longer are a worry but there other new things you do need to worry about. This section will outline each of these items and the issues related to them.

4.4.1 What Fluid Should I Fill the Sea Chest With?

If you have not placed holes in the window and you are not going to work in an area where freezing is an issue then the sea chest should be filled with fresh water. Fresh water decreases the issues of corrosion in the sea chest.

P/N 95A-6019-00 (December 2002) page 17

If you will be in an area where freezing of fresh water would be an issue then seawater can be used.

Some users have placed ethylene glycol into the fresh water well to prevent freezing. Although this will not harm the transducers you will have to per-form post processing on the data sets from the NB, BB, and WH ADCPs (this issue is not present for the OS ADCP). The NB, BB, and WH ADCPs must have the velocity data scaled properly based on the speed of sound in the sea chest. Ethylene glycol causes the water to have an inverted speed of sound change to that of fresh water or salt water. This means that RDI’s standard software programs will not be scale the data properly. You will have to record separately the speed of sound in the sea chest and then in post processing correct the ADCP velocity data appropriately. However,

4.4.2 How Much Fluid Should I Use in the Sea Chest?

The transducers must be completely immersed in water. No air should be in front of the transducers and the pressure within the sea chest should be ad-justed to keep the window from bowing in and out, and thereafter, the vol-ume should be kept constant.

4.4.3 Should I Use Absorption Material When Using a

Window?

The window causes some of the transmit signal to reflect back into the well due to the difference in impedance between the window and the water.

When the transmit signal is reflected in the well it becomes trapped and this results in what is called ringing. To keep from processing this signal, the blanking of the ADCP will have to be increased.

However, in extreme cases, ringing can last a period that will cause the first 50-100 meters of data to be unusable. Therefore, a sound absorbing material should be used inside the sea chest to minimize the effects of sound ringing within the sea chest. The material should be a minimum of one wavelength thick (include the sound speed of the absorbing material when calculating the size of a wavelength). Approximate wavelengths of sound in seawater are given below in Table 3. Using standard neoprene wet suit material has been found to work well with 75 and 150hKz frequency ADCPs.

Table 3: Wavelength of sound in seawater (1500 m/s sound speed) Frequency (kHz) Wavelength (mm)

38 40

75 20

150 10

300 5 page 18 RD Instruments

4.4.4 Do I Need to Worry About Corrosion When Using a

Window?

Corrosion is always possible. However, our transducers are made of a ma-terial that has shown to corrode very little over time when the above precau-tions are met. There is nothing that you can do to protect the transducer from corrosion. However, if the well is covered with a window and then filled with fresh water corrosion is can be further minimized. You should inspect the transducer regularly for signs of corrosion.

Note, never attach any anodes directly to the transducer head. Additional anodes or impressed voltage systems can cause the urethane to separate from the transducer (cathodic disbondment) or cause the material of the transducer to break down. Standard anode protection used for the ship should be installed outside of the well of the transducer head. Mounting of ship’s standard anode protection outside of the transducer well will typi-cally not cause any problems.

4.4.5 Do I Need to Worry About Air Bubbles When Using a

Window?

All vessels create air bubbles in the water as the ship moves through the water. Ships with a deep draft or a non-flat bottom have fewer problems with bubbles. If you are using a window these bubbles will still be present.

If the window is sealed then this air will not fill the sea chest. However, if the window is not sealed then air can fill the sea chest. You must make sure to vent air from the sea chest periodically if there is a possibility that air will become trapped in your sea chest.

To avoid air bubbles from getting into the front of the window you should mount the transducers below or away from the bubble layer. The flow layer is usually within the first two feet below the hull. Bubbles can be trapped in this layer. Mounting the transducer head amid ship on the fore-to-aft centerline may help. Another technique is to divert the bubble layer so it flows around the transducers. A fairing around the sea chest can help with this, but care must be taken so that you do not cause cavitation.

4.4.6 Do windows Improve Flow Noise Problems?

Water flowing over the transducer faces increases the acoustic noise level, which decreases the profiling range of the ADCP. A window reduces the coupling of flow noise to the transducer. This is because of the gap filled with fluid between the inside of the window face and the transducer faces attenuates the flow noise. By reducing flow noise you are increasing the signal to noise ratio. The higher the signal to noise ratio the better the stronger the returned signal will appear. This will result in better data re-ception and longer ranges.

P/N 95A-6019-00 (December 2002) page 19

4.5 What Maintenance is required when using

Windows?

In general, a window provides protection to the transducer from the most common sources of problems such as bio fouling and corrosion. However, the window can still become covered with bio fouling or could become damaged. The following section discusses these issues.

4.5.1 How Often Should I Inspect the Window?

Since growth of mussels, barnacles, and other bio fouling occurs on all ves-sels the window should be inspected and cleaned by divers on a regular in-terval. This interval should be often enough to prevent the growth of any-thing on the window and to allow inspection for damage to the window. It is recommended that this interval be at least once per year, but may be re-quired more often in areas that have heavy bio-fouling growth.

When inspecting the window you should inspect for bio-fouling growth, cracks, damage, for air pockets, and for mud on the inside of the window.

We have seen cases where the inside of the well became filled with mud.

The mud entered through a crack in the window and where the holes were drilled in the window. Bio fouling should be cleaned off, air should be purged from the sea chest, and mud should be removed from the sea chest.

4.5.2 If I find that the Window is Damaged Can I Keep

Operating the ADCP?

In general, any window that is cracked or is damaged so that it is not smooth should be replaced as soon as possible.

A window that is damaged causes a problem with the acoustic transmission.

The exact problem or problems seen because of this damage will vary de-pending on where the break is and the way a beam would strike the damage.

All windows have losses because of an impedance difference to the water inside the well and outside the well. There are also losses that are built up in the window. An important loss is due to the shear wave that is created as our acoustic signal passes through the window at an angle. This shear wave traps sound in the thickness of the window as the acoustic signal tries to pass through the window. If the window has a crack in it then the window can cause this trapped energy to bounce in all directions rather than remain trapped in the window. Depending on the size of the crack, the location of the crack, and what the window does around the crack this reflected energy may even go into other beams.

Regrettably, there is no way to predict on what can happen as a crack will have a strange pattern to it. A single beam or all four beams may be af-fected. However, in either case it is enough to know that a crack in the page 20 RD Instruments window is very bad and will cause the energy that is transmitted and re-ceived in a beam to be deflected at strange angles.

Additionally, cracks can cause the window to have a rough surface. This can result in cavitation around the window. Cavitation results in air being trapped near the crack. This air can cause energy to be reflected back into the transducer well instead of traveling through the water.

4.5.3 Does the Use of a Window Effect My Warranty?

The use of a window has no impact on warranty. The window is primarily an aid to optimal performance. A window isolates the transducer face from flow noise when the vessel is moving and provides protection from bio fouling. These all increase the performance and reliability of the trans-ducer. The window will also absorb some of the transmitted and returned signals. This will have an adverse effect on performance. However, when the proper window is used this adverse effect is minimal compared to the benefits of using a window. RDI cannot be responsible for the acoustic de-sign of the vessel, but that design and the installation of our transducer cer-tainly can adversely affect the ADCP system performance.

4.6 Insertion Loss

The following section contains insertion (one-way) loss graphs for each of the ADCP frequencies at each transducer beam angle at 0°C. These graphs are provided as an example of the expected insertion loss.

The main beam of each ADCP system at its maximum bandwidth is dis-played as the red line on each graph (the sin X/X is represented by the smaller bumps in red). The minimum and maximum frequencies used on the X-axis of the graph were chosen so that this bandwidth would be ap-proximately centered on the graph.

The blue line represents the expected loss across this bandwidth of frequen-cies for this thickness of polycarbonate. The Y-axis of each graph repre-sents the expected insertion loss. See the example below for descriptions.

Uniform Ave. IL represents the entire average insertion loss over the entire frequency (X-axis) shown. The Weighted Ave. IL represents the average insertion loss over the bandwidth of the ADCP frequency. The Weighted Ave. IL is used to complete Table 1, page 10 and Table 2, page 11 of this document.

P/N 95A-6019-00 (December 2002) page 21

Figure 4. 38kHz Insertion Loss (one-way) with a 19.1mm window at

0°C

Figure 5. 38kHz Insertion Loss (one-way) with a 38.1mm window at page 22 RD Instruments

Figure 6. 75kHz Insertion Loss (one-way) with a 9.5mm window at

Figure 7. 75kHz Insertion Loss (one-way) with a 9.5mm window at

P/N 95A-6019-00 (December 2002) page 23

Figure 8. 75kHz Insertion Loss (one-way) with a 19.1mm window at

Figure…