How narrow should an echosounder beam be for a small ROV or USV?

Beam width is a cone, and the cone gets wide fast

An echosounder does not measure the distance to a point. It insonifies a cone of water and reports the distance to the strongest echo coming back from inside that cone. The footprint on the seabed is a circle whose diameter grows linearly with altitude:

footprint diameter = 2 × altitude × tan(beam width ÷ 2)

Applied to the two beam widths common on small-vehicle echosounders:

Calculated from published beam widths. The ratio is constant: 5.1× in diameter, roughly 26× in area, at every altitude.

At 30 m over a harbour floor, a 25° device is averaging everything inside a 13 m circle and giving you one number for it. A 5° device is averaging a 2.6 m circle. If there is a 3 m boulder down there, the narrow beam can tell you the boulder is a boulder. The wide beam tells you the bottom is somewhere between the boulder and the mud around it.

This is why beam width, not range and not depth rating, is usually the specification that decides whether a given echosounder suits the job.

The options for a small ROV, AUV or USV

Two devices cover most of this market, and they sit deliberately at opposite ends.

Blue Robotics Ping2 - 115 kHz, 25° beam, up to 100 m range, 300 m depth rating. It arrives as a sealed hard-anodised aluminium unit with a 1 m cable, a WetLink penetrator already fitted, and a mounting bracket in the box. Open-source software interface, Arduino, C++ and Python libraries. Blue Robotics describe the 25° beam as suited to a rocking boat, which is an accurate description of what a wide beam is for.

Cerulean Sonar Sounder S500 - 500 kHz CHIRP, about 5° beam, 0.3 m to 100+ m range, 300 m depth rating. It arrives as a 69 × 38 mm circuit board and a separate 54 mm transducer. The board goes inside the vehicle’s existing pressure housing; only the transducer is wet. Ethernet, USB or 3.3–5 V serial.

Cerulean Surveyor 240-16 - the answer when neither single-beam option is enough. 240 kHz, 16 receive channels, 80° cross-track swath, under 1° angular resolution. Not a single beam at all, but the honest next step, and worth knowing the price of before committing to a single-beam survey plan.

EUR prices are SepcoTech’s own, excluding VAT and shipping.

What 500 kHz and CHIRP actually buy you

The narrow beam is a consequence of the frequency. A higher-frequency transducer of a given physical size produces a narrower beam - that is why the S500 gets about 5° out of a 54 mm face at 500 kHz, and the Ping2 gets 25° at 115 kHz. You do not choose the beam width independently of the frequency.

CHIRP is separate and does something else. Instead of transmitting a single tone, a CHIRP device sweeps the frequency across a band - 20 kHz, in the S500’s case. Correlating the returned sweep against the transmitted one localises the echo in time far more sharply than a single tone can. That is where the banded range resolution comes from: 3 mm out to 17 m, 9 mm out to 50 m, 24 mm beyond that.

Note the bands. A figure like “3 mm range resolution” is true at 17 m and not true at 60 m, and any supplier quoting it without the band is quoting a headline rather than a specification.

The higher frequency costs you absorption. 500 kHz attenuates in seawater considerably faster than 115 kHz, which is why a 500 kHz device reaching 100+ m is doing well rather than doing the obvious.

The honest caveat: ping rate falls with range, and the S500 is a component

Two things buyers discover in week two.

Ping rate is bounded by the speed of sound, not by the electronics. The S500’s specification sheet says 10 pings per second. Cerulean’s software documentation is more specific: 10 pings per second holds out to about 65 m, falling to about 7 pings per second at 100 m. Enabling CHIRP can reduce it further in some circumstances, because of the computational overhead of the correlation.

This is physics and it affects every echosounder ever built. A ping cannot be sent until the previous one has travelled to the set range and come back. At 100 m that round trip is roughly 0.13 seconds in seawater before any processing. No product design escapes it. What it means in practice is along-track sample spacing:

For almost all shallow survey work this is more than sufficient, which is the point - the falling ping rate sounds alarming and usually is not. Worth checking against your line spacing before you commit, not worth worrying about.

The S500 is a component, not an instrument. This is the one that catches people. There is no enclosure, no bracket, and no penetrator in the box. Cerulean recommend an M10 penetrator for 6 mm cable, bought separately, and their documentation walks you through prising open the JST-GH connector with a hobby knife, pulling the three wires out individually, passing the cable through the penetrator and re-terminating. The spare connector in the box is there because that operation sometimes ruins the original.

There is an alternative - seat the supplied O-ring in the groove on the back of the transducer and take the cable straight into a waterproof enclosure - but it constrains where the transducer can sit.

None of this is a defect. It is the direct cause of the low drag and the 50 g wet mass, and for anyone building a custom AUV it is a feature, because a board and a transducer fit where a sealed cylinder does not. But if your mental picture is bolting a sensor to a thruster guard and having the vehicle find it, that picture is the Ping2.

One more integration detail worth knowing before the first power-up: after each boot the S500 polls all three interfaces, and the first one to send a valid packet becomes the host while the other two are disabled. Bench-testing over USB and then deploying over Ethernet needs a power cycle, not a cable swap.

Do I need a DVL as well?

This is the question that follows, and the answer is usually no, for a reason worth being clear about.

An echosounder measures the distance to the bottom. It tells you nothing about where you are horizontally. A Doppler velocity log measures velocity over ground and lets a vehicle hold station or dead-reckon a track.

If you want altitude hold, an echosounder is enough. If you want position hold, waypoint navigation, or geo-referenced survey data, you need positioning - a DVL such as the Cerulean Tracker 650, or an acoustic positioning system, or GNSS on a surface vessel. Bathymetry in particular is not a depth problem, it is a depth-plus-position problem, and a sounder on its own produces a list of depths with nowhere to put them.

Deciding

Choose a 25° device if the job is obstacle avoidance or altitude hold on a vehicle that moves, if you want it mounted and working the same afternoon, or if the platform will pitch and roll enough that a narrow beam would lose the bottom.

Choose a 5° device if the depth figures are going into a chart, if you need to resolve features rather than average them, if weight and drag are binding constraints, or if you are integrating into a custom vehicle where a bare board is an advantage.

Choose multibeam if you need coverage rather than a line of soundings. A single-beam sounder gives you depth directly beneath the vehicle and nothing either side of it, and no amount of narrow beam changes that.

Frequently asked questions

Talk to us before you order. SepcoTech supplies both the Cerulean Sonar and Blue Robotics ranges, so we have no reason to talk you into the wrong one. Tell us the vehicle, the water depth and what you intend to do with the data, and we will name the part number - and say which one we ruled out and why. Clear EUR quotes, brackets and cabling included, support in your time zone.