8
\$\begingroup\$

I am working on a project from University which includes a ROV. I am trying to come up with an easy way to create a Distance Sensing system, which doesn't require a lot of work. something very simple that can measure up to 20 cm, with a fair resolution.

I have looked into ultrasonic but apparently it doesn't work well in water + the waterproofing is a hassle, also Infrared is not ideal either. I purchased a fish finder so I can use the Transducer but that has a lot of obstacles too, i.e creating filters.

Any easy way that I can hack a fish finder, or create a simple sensing unit using a MCU and a sensor?

\$\endgroup\$
  • \$\begingroup\$ I looked into this for a university project as well and fish finder was what we concluded at the time, never got to implement it though. \$\endgroup\$ – Joe Baker Jan 22 '13 at 22:04
  • \$\begingroup\$ The longest distance you want to measure is 20cm. What's the shortest distance that you want to measure? \$\endgroup\$ – Nick Alexeev Jan 22 '13 at 22:28
  • \$\begingroup\$ It would need to be calibrated depending on the water each time, but I wonder how two electrodes arranged like /\ but insulated at the top would work. Just checking some tap water and resistances were in the one megohm type range, but depends on if something sticking out the top is OK. If that's OK a small buoy with a rotary encoder and some sort of mechanical tensioner might be something else to think about. \$\endgroup\$ – PeterJ Jan 23 '13 at 3:17
13
\$\begingroup\$

If the requirement for hacking a ready-made device can be dispensed of, a low cost and simple to implement range sensor option is:

  • Use a 405 nanometer blue laser as a projection element. These are available for around $13 on eBay: Blue-violet Laser Module Use epoxy potting compound to waterproof the module - the lens mounting is already waterproof, don't use epoxy on the lens.
  • Alternatively, modify an even cheaper (less than $5 on eBay) 405 nm 5 mW laser pointer to waterproof it and pull power leads out. You can try the finger of a transparent rubber surgical glove for a workable waterproofing option.
  • For extended coverage: Split the laser beam using a parallel prism - a 5-way split would be ideal for 5 beams in a flat plane.
  • Use a phototransistor or photodarlington as a sensing element - choose a part with good sensitivity at the blue end of the visible spectrum
  • Mount the sensing element close to the laser, pointing in the same direction as the laser.
  • To improve selectivity, put a violet-blue gel filter in front of the phototransistor to block or significantly attenuate incident light which is away from the desired (405 nm) part of the spectrum.
  • As the laser + sensor combination approaches any reflective or dispersive surface, the intensity of blue (405 nm) light incident on the phototransistor increases. This occurs even if there is no specular reflection from the surface, because the collimated laser beam(s) still gets reflected back in the source direction due to speckle pattern formation. Laser proximity sensor
  • Unfortunately this approach will not work if the water has significant bubble activity, or suspended particulate matter, since the speckle pattern intensity will remain high even without a solid reflecting surface ahead.
  • Despite this sounding a bit counter-intuitive, the reflected signal intensity in water of a 1 mW blue laser is typically strong enough for detection by a phototransistor even though there is no perpendicular reflecting surface on which the laser is incident - whereas a direct reflection is likely to overwhelm the phototransistor, so appropriate precautions are needed.
  • Calibrate the received signal intensity on the phototransistor for distance, and you have your underwater proxmity sensor.
  • Don't forget to use epoxy potting to waterproof all exposed circuit traces, components, wires and connections.

Why this works where infrared LEDs or lasers wouldn't:

  • The red end of the visible spectrum (and more so the Infrared) is absorbed by water about 100 times as strongly as violet/blue: Light absorption spectra in water from this article

Why this works with a laser but not necessarily with UV LEDs:

  • The light from an LED is not collimated, hence does not get significantly intensified in the return direction, whereas a combination of reflection and diffraction causes significant pulsed intensification of such returned signal for a collimated laser beam, by speckle pattern formation.
\$\endgroup\$
  • \$\begingroup\$ +1 Really cool answer. I wonder if using a red/near IR laser would be a better choice since the OP is only concerned with a very small distance and the water's absorption may help reduce interference, especially the deeper it is under water?? \$\endgroup\$ – Garrett Fogerlie Jan 23 '13 at 15:48
  • \$\begingroup\$ @GarrettFogerlie Good point... and 980 nm IR lasers are easily available, as are IR photosensors, including some pretty sophisticated integrated parts with great IR-pass filters built into the lens. \$\endgroup\$ – Anindo Ghosh Jan 23 '13 at 15:55
  • \$\begingroup\$ Thanks Anindo I'll give this a try. Any suggestion on the phototransistor ? \$\endgroup\$ – user1622997 Jan 23 '13 at 17:07
  • \$\begingroup\$ How can you calibrate this system when the nature of the reflector is unknown (the OP is using a ROV which is moving through the water and hence its local environment will be constantly changing)? Relying on intensity variations to measure distance is very risky. With such a small distance, ultrasonics should work as the attentuation will be very low. The pulse width required depends on the desired resolution which is not given. \$\endgroup\$ – Barry May 6 '17 at 21:13
3
\$\begingroup\$

This Underwater Ultrasonic Sensor for distance measurement may be just what your looking for. As Michael Karas pointed out, that ultrasonic sensor is rated from 30cm-3m, sorry. However, this site may have another suitable option.

You could create your own module with an underwater transducer. You should read this if you haven't already.

Any easy way that I can hack a fish finder, or create a simple sensing unit using a MCU and a sensor???

I'm not very familiar with fish finders, but I would bet you could get a real cheap fish finder that has a data output and try to mod that to work with your project. However I think the range will be to large for you, since you only need 20cm, a fish finder (or it's transducer) will most likely not be too accurate at this distance.

This is the cheaper version of the first module I linked to, (at least I think it is) Waterproof Ultrasonic Sensor Distance Measuring Module is not meant to be used under water but it may work for you depending on the depth of your device.

\$\endgroup\$
  • \$\begingroup\$ The first transducer/range finder that you linked is only good for making measurements down to 30cm. That is not in the ball park of the OPs 20cm max distance. \$\endgroup\$ – Michael Karas Jan 23 '13 at 1:39
2
\$\begingroup\$

I have run into this water depth measurement before and found that a pressure sensor and a bubbler (i.e., a fish bait tank airator) works nicely as demonstrated and explained in this Floating Dock Patrol YouTube video https://youtu.be/0CRarPCHXk0 . The micro in the Floating Dock Patrol is one from Parallax (Raspberry Pi would work too). I find Mr. Ghosh's answer very interesting and I am going to try it (I'm in the process of procuring the parts for the laser idea). I tried hacking a fish finder but the accuracy of a fish finder in shallow water is poor and lots of noise. Ultra sonic sensors don't work as they just bounce off the water. The bubbler works the best so far but the blue laser might be a simpler approach.

\$\endgroup\$
1
\$\begingroup\$

I may be a little late to the party on this one, but any of the sharp IR proximity sensors will work underwater and are ideal for short ranges. I use them for this very application on the project I am currently working on. You will need to waterproof them and place them behind some sort of IR permissive case. I use a GoPro housing with the sensor fixed in place and a small through hole gland installed (can be sourced from McMaster Carr or similar supplier). They do have a non-linear response and you will likely have to map values to the response of the sensor based on testing as IR attenuates pretty heavily in water, but this sensor is low cost and is incredibly easy to install and leverage. Of note, it is only ideal for short range applications.

\$\endgroup\$
  • \$\begingroup\$ Of note, the through hole gland is for the wire which suppllies the voltage, ground and signal wires. \$\endgroup\$ – Matthew Connolly Nov 7 '18 at 15:38
0
\$\begingroup\$

As sort of an empirical reference, my experience with various fish finders suggests they do not work well in less than 1ft of water. I suspect they need a minimum distance to properly hear the reflection. If the transducer is up against a flat surface, the readings are incorrect. It is also possible the receiver is highly directional, so when the receiver is <1ft from the reflection surface, it misses the reflection.

\$\endgroup\$

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.