Underwater metal detector


Hi everyone,

I just bought the ORov 2.7, and my imagination started to fly.
Where I live (Menorca) are plenty Roman wrecks, from 2000 years ago, also turkish pirate wrecks. About roman wrecks, only the anchors could be detected, pirate vessels also the canons.
The thing would be in conjunction with the WiFi Buoy, make a good inspection of every area I want to inspect. Leave the rov underwater, anchored to the buoy, and only take it out for battery replacements.

Do you think this is possible?

I saw some arduino metal detectors, this is not what concerns me. I’m only worried about the power I would need to detect some metal in the water and at 20cm or 30cm from the bottom.

Thank you!

Metal Detection

In all respects, yes, this is one of the concepts the OpenROV design allows for. Watch the Long Range buoy thread as we develop that, but the concept already works and you can certainly make one up right now.

The OpenROV community hasn’t delved into metal detector payloads just yet, but it is certainly one of my eventual requirements for a project. To get the distance an penetration you are looking at, I think a traditional metal detector might be a bit short. I have been looking into a Pulse Induction (PI) detector, but I haven’t figured out how I want to monitor returns. I was thinking of a an external light bar next to the camera.

Anyway, here’s a PI system that I have been looking at: Aquapulse AQ1B


Hi, I was playing with surf 2.1 pulse induction kit, for mi rov ,no good result yet ,a year a go i was focus on edge EMD-2 http://subseaworldnews.com/2013/08/20/edge-launches-new-metal-detector-emd-2-uk/
but apparently edge web page is vanish, http://www.edgedesignlab.com/


An intermediate technique is to monitor the field effect on the magnetometer data.


And cheaper I guess. I thought about that, but I’m afraid about assembling the magnetometer right under the rov, I think it would give me corrupted data if it’s to close to the electronics


Yeah, there’s that as well. The kit’s size has its limitations. Now @Kevin_K kit is a bit more multimodal and would allow for expansion and tool sleds, etc.


My take on using OpenROV as an underwater Metal detector is that it could (should??) be done. It may be somewhat better on something similar to @Kevin_K 's beast but if it is being used in a not heavily current prone area it would work on a less modified unit (some still obviously needed)

I personally do not think a MEMS (IMU) based version would have much if any discrimination and would have to be so hard against the Iron/Steel object to get a decent result that it would be unworkable (the earth is a very big magnetic field to see smaller ferric items against that background) additionally

Although remember both Pulse Induction (PI) and Very low frequency (VLF)(AKA induction balance) with a coil fixed say 600mm below the ROV would both potentially work (advantages and disadvantages to both styles) and neither are a magnetometer (detecting ferric items but are metal detectors basically sensing conductive material). The amount of metal in the ROV can almost be considered as a background noise and essentially calibrated out

If the coil was negatively buoyant below the unit you could also add some buoyancy to the top of the unit give it greater righting moment / stability in it


@Scott_W, Hi Scott, I don’t completely disagree :smile: however the MEMS magnetometers are actually quite sensitive to local disturbances and most Nav systems employ filtering techniques to mitigate local EMI and other disturbances. We’ve had students in the past use the mag data on land rovers to map local effects (field effects) in buildings rather effectively. Localized fields tend to mask the Earth’s field. The Earth’s field is actually somewhat weak. Something like 60 microtesla. That’s why you want you magnetometers as far away from any EMI in your system as possible, otherwise you mask out the Earth’s field. The MEMS mags are not ideal, but it would be an interesting experiment on the ROV, nonetheless. (non-ferrous materials are invisible). But please correct me if I am wrong.

I do agree with the coil solution for detecting metals in general, active system as it were, as the Mag is passive and only gives you static/dynamic field reading and can not induce them.

Anyway, best way forward is to pick up a metal detector from a local yard sale, and slap it on the ROV. :smile: I’m rather surprised someone hasn’t done this yet…


@Jim_N I would have no problem installing a large metal detector on my MSOC ROV. I can figure out something for power. The problem is one of how do we interface it? How would I, as the operator, know when I drove over something? Most marine metal detectors still have a pair of earmuffs for the diver.

@Scott_W Just curious if the signature from the ROV can be calibrated out, why is it not more readily done? Almost all the ROVs I’ve seen with mounted metal detectors have it on a boom, far away from the ROV.

Something else I was thinking for a payload for stock OpenROVs, was something like the Garrett Pro-Pointer AT taken apart and put in an OpenROV Battery tube. The range probably won’t be that great, but it’s the smallest that I think will fit on the OpenROV frame.


Neato. No worries about the signal. I’ll dig into it, but I suspect its just an A/D converter to some logic for the tone intensity that we can map to either alphanumerics or color indications on the OpenROV HUD. I would image it as three or four concentric rings, each one being color coded given strength of the signal, and visually representing if you are on top of the source. i.e.

Now for directional capability, we’ll have to sus that out…


Oh, and Scott please correct me if I am wrong, but the detectors induce eddy currents in metals that they pass over, whereby creating additional current in the metal, thus allowing for the field detection, so if your ROV isn’t under the coil, there shouldn’t be any current induced.


@Jim_N Wow, Ok, if you can get something hooked up for the HUD, that would be really cool. If not the rings, just a simple signal strength bar should do it. I was thinking something like below if I was going to have the indicators external. But if you can make a plugin for the HUD that would be much better.

I don’t think we’'ll be able to figure out the directional, that’s why metal detectorists have to wave it all over the area right?


No Jim you’re not wrong, it’s all about those eddy currents. I’ve played a little bit with the hand held U/W metal detectors and they seem to be OK to say 300mm below a sandy surface for coins etc. But I’ve had a lot more experience with proton mags looking for wrecks.

Given the mantra of tiny steps forward – all good.------------But in the bigger picture my take is to try and get it back as a reading into the unit, so some time in the future, it can tie back to what @Jim_Trezzo is doing with the acoustic positioning

and what @Jim_N is doing in relation SLAM

so a bigger picture “heat map” may be able to be produced for a “site”


Hi Jim, Kevin,
Any updates on this? I would also be interested in adding a metal detector on my ROV for search and recovery in the bad visibility in (city) waters of the Netherlands. In absence of sonar… I have done some googling on arduino metaldetectors and am considering to make a 20cm spool with about 30 windings to create a magnetic puls and pick up the response in a response spool. Since I am complete new to detectors, I have no clue how this will work under water. But I will keep you posted on any progress (may take to winter :slight_smile: ). I am also interested in ways to read the coil output and send it to the surface in sensible language. Maybe an audio input needs to be added to my ROV to send raw data up: no clue yet. Or another amplifier to get amplitudes from the spool.


Hey there Niels,
I’ve been working more on the ROV itself than payloads. I’m about 6 months off from anything more than cameras.

Please continue your work and share what you come up with here, I know we’ll all benefit.

My requirement will be to detect 2-3 kg of iron under about 10cm of sand in 60-75m of water.


Hi Niels,

Unfortunately no. I’ve so many freaking projects, it’s untenable, lol. Should get into it in the next few months though. Let’s keep the thread going.


:+1: +1

Yep so much happening uuurrgg the day job’s killing all of the important stuff


Most beginners get excited by prospecting gold. However, keep on mind that gold field is typically located in areas with higher mineral levels. These ground with minerals actually refers to the magnetic properties of the soil. The minerals on the ground don’t necessarily produce the magnetic fields, but it only means that the soil grains and particles can be attracted to magnets. By using metal detectors, it picks up signals from the mineralized ground as a target, which makes it difficult for you to detect the buried items in the mineralised ground. You can solve this problem by using specific a gold prospecting metal detector with superior ground balancing system that counters the ground mineralization.


Is there any guide for underwater detectors as full as this one http://tenrows.com/metal-detectors/ for casual detectors?