| by Huffman | 2 comments

The module of the underwater ultrasonic rangefinder

This device came to light in the process of working on a computer project for diving, which in its composition was to have escaladieu sonar rangefinder. Whether we are bad looking, or, indeed, on the market, nothing suitable for the options sold, but in the end, instead of purchasing a ready-made module we decided to dive computer to invent your rangefinder. Moreover, the process of integration into the compact body of the computer seemed easier when you post the emitter and the electronic circuit in accordance with the free space inside the enclosure.

While the dive computer is waiting in the wings, we came up with the idea: what if the meter is in a separate project? Because we had difficulty finding a suitable module of an underwater rangefinder, surely someone else has experienced similar difficulties search.

The initial requirements for the module were the following:

  • a simple digital interface. Ideally UART with text commands/responses.
  • voltage – 5 Volts
  • most compact, waterproof housing

In order not to delay the narrative, I cite from the initial working schema with some explanations in the text.

image

As you can see, the scheme represents the classic version, is divided into three main blocks:

  • Managing
  • Transmitting
  • Host

As a control, was chosen MK STM32F302K8U6 as a good combination of price, performance and size. The area of responsibility of this unit is to generate a signal to the sending unit, to digitize and process the signal from the host, and custom applications to grant access to configuration and measurement data in a convenient format.

Sending unit contains: gate driver, power switches, matching-up transformer, anti-parallel Assembly of diodes. The last element, generally speaking, not mandatory in such schemes, but in our case, plays a vital role in “turning off” the transformer at the stage of acceptance, when the value of the current voltage on the piezoelectric element below the opening voltage of the diode, which helps to isolate the input stage of the amplifier of the received signal from the transformer.

The host block is the most demanding to the quality of execution, from which depends almost everything. On the one hand, needs to absorb hundreds of volts at the input of the stage of generation of the probe pulse. On the other hand, to provide enhanced signal amplitude of the order of microvolts with an acceptable ratio of signal to noise and to the level of confident detection and capture of the ADC. For the ability to remain unharmed when hundreds of volts at the input corresponds to two components: high-voltage capacitor C2 on which the AC and antiparallel Assembly of diodes which limit the voltage to acceptable values. It should be noted that this circuit is able to perform its task only in a pulsed mode with duration of the order of tens of microseconds.

For signal enhancement answer three stage active bandpass filters, of which the last two are narrow-band filters of the second order, providing a good selectivity in frequency. However, the price of a good selectivity in frequency is a more rigorous approach to the selection of errors for castorocauda components, or do not provide the stability of the parameters from instance to instance. In the framework of the budget devices are optimal for our problem turned out to be the error of 0.1% for resistors and 0.5% for capacitors. Further, according to the scheme: amplitude detector and a passive low-pass filter to suppress carrier remnants that passed through the detector.

As for the acoustic parts of the device, which is not visible on the concept, here the basic element is a piezoelectric disk GGC.757681.008-172, with a diameter of 12.9 mm, thickness 3.1 mm, a resonant frequency 640кГц. This element is both the emitter of the probing pulse and receiver of the reflected signal. From the aquatic environment it is isolated quarter-wave matching element (polyurethane resin). Together with the analog circuit, the design of emitter/receiver makes a great contribution to the quality of the whole device. The speed of sound in the plastic, which is used as a matching emitter/insulating element is 2400 m/s. This determined the thickness of the matching element is 0.9 mm.

image

In the prototype of the PCB there were some shortcomings, but surface mounting we made this device operate at a minimum acceptable mode. That was enough to characterize the radiation pattern.

image

As a bench for measurement of the direction we haven’t thought of anything better than to use a plastic trough full of water where you installed our module-meter as a radiator, and a short distance from his receiver a piezoelectric element connected to the oscilloscope. Module-meter can be rotated left-right from 0 to 180°, at this time receiver we remove the waveform and determine the amplitude of the received signal depending on the angle of the emitter.

The following is a schedule of the final radiation pattern.

Further measurements were carried out in a makeshift pool – a plastic tube filled with water to a height of 1 meter. By the way, it was possible to detect three reflections from the bottom and the water surface, before the signal disappeared entirely.

Since we decided to use a UART interface with simple text commands (the first stage), it is not big deal to implement a bunch of module-finder with GUI on the computer for Processing.

Photo — scan of the bottom of the local pond by using boat mounted with a module of the range finder and a radio transmitter for receiving data from the module to the laptop. Unfortunately, in moments of emotional trials glands, least of all think about the quality of the photos of the process.

And using the smartphone app, Bluetooth Electronics and the Converter USB->UART can be implemented more mobile variant of the test setup.

At the moment we are working on the next prototype of the module of the range finder as to improve, in fact, is what.

For example, to realize a gradual amplification of the input signal, so that the reflection from nearby objects rolls over, and from the distant objects enhancement was maximum. Maybe we can use cheaper circuitry. Consider the option of filling the inner space with some kind of compound or silicone oil to seal and confronting great pressure.

2 Comments

Magawaik

Feb 2, 2019, 2:26 pm Reply

What is the processing? You just draw the amplitude? Convolution with a perfect hump makes no sense to do?

admin

Mar 3, 2019, 2:26 pm Reply

Eliminates the noise of the ADC through the digital low-pass filter, then determines the maximum amplitude in the given time window is the graph element. It turns out that Yes, just the amplitude. About a hump is it?

Leave a Reply to admin Cancel reply