Water Sampler



During the NASA International Space Apps Challenge, one of our goals was to build a water sampling device that could be attached to the payload bay on OpenROV. Water sampling is a very important capability for OpenROV to have in many situations because it allows researchers to bring water from the area they explored back to the surface where chemical, biological, and particulate analyses that may not have been possible in-situ can be done.

The way a water sampler works is quite simple- it allows water to flow into a chamber freely, then when a sample is to be taken, that chamber is closed so the water can't escape. This way, if a sample is taken it won't be contaminated from water at other depths as the sampler returns.

The concept we came up with was to have a tube which would be open at both ends mounted onto the payload bay with an extension spring running through the center of the tube which would pull to plugs together upon the release of a trigger.

Here are some of our initial sketches.

We went to the hardware store and picked up a few bouncy balls that would work as plugs and built a proof-of-concept model which seemed to work. When we cut the piece of fishing line holding the balls out of the tube, they pulled together to seal the tube.

We then developed a way to do the same thing with rubber stoppers (which would presumably make a better seal) and we built a mount that could be attached to the 50mm space of the payload bay rods. We also began considering some of the issues involving the triggering mechanism, namely that if a string were to be cut, the loose ends could get entangled in the props of the ROV. Our solution was to make a tensioner which would also function as a mechanism for giving compounding mechanical advantage at holding the plugs apart.

We were happy to find that the second version of the sampler also worked quite well and didn't seem to leak at all! After mounting the sampler to the payload adapter (we designed it to mount with zip ties so it could be easily removed and replaced with a fresh sampler during a mission), we patted ourselves on the back and moved on to the next challenge- designing the trigger mechanism. Any ideas?

*Special thanks to all the people at the Space Apps Challenge who helped us with the development of this awesome device!


I would go with either a servo (a lot of work), and just remote control it or making a lever that extends past the front of the ROV and crashing into something lightly to activate it...lol


On facebook Matt Crenshwaw wrote....

If you're looking for a release for around $5/pop or so I would use an epic trigger resistor and a small amount of model airplane parachute ejecting type powder to pop...something. It doesn't use very much power to trigger that kind of release, it's cheap, and it's reliable. We use that kind of release to pop a nylon bolt to get our animal tags off animals and you could probably get our primed epic trigger resistor PCB's from us. You can adjust down the amount of powder so it doesn't blow the whole thing up.


An electromagnet, that is inside the main tube. If the trigger mechanism is sensitive, the magnet doesn't need to be so powerfull.


There is a parameter that you didn't take into account regarding the sampler mechanism: pressure.

If the tube of the sampler starts closed (to be kept clean for example), there is air inside it trapped at 1atm pressure. When you'll need to open it let's say at 30m/100ft deep you'll need to counter the extra 3atm pressure of water.

To solve this, you could reverse the logic of the trigger: The sampler tube could just start open, letting the water flow right through it and you can just close it with the trigger when you are at the target location. This could introduce some contamination of the final sample (algae, trash etc) trapped inside the tube while it was open. Another solution is to fill the tube with water before you close it at the surface, either sea water taken on site or even desalinated (I can't remember the proper term) water so as to equalize the external pressure (remember that water is practically uncompressible).


The tube is designed to stay open until triggered when it is closed. This way, water during decent can flow through the containment area (which could introduce contaminants) but the assumption is that this would be used for samples where that amount of fidelity doesn't matter. In cases where no contaminants from other water levels are permissible, I've considered a sort of spring-syringe design that would cause a syringe to draw water when triggered. Surrounding water (that would not end up in the sample) would be on the other side of the syringe plunger to keep displacement the same and would be pushed out of the way by the drawing action. A second mechanism that plugs the inlet to the containment area after it has drawn would also likely be needed to prevent contamination during ascent back through the water column.


Not fully thought out, but you could use a syringe with the needle end blocked. As the ROV descends, the ambient pressure will push the plunger to compress the air. The distance the plunger travels be used to trigger the device.

Issues would be calibration, stiction of the plunger, and visual confirmation that the device released at the target depth range.

I had a use for this two weeks ago, chasing water samples in a water column 6m deep with more than one halocline.


Sorry if I am being stupid, I'm no engineer but an IT geek .. But why the complex design? Wouldn't it be easier to use one or two small motors and a "sliding door"? You wouldn't have to worry about the water pressure (yeah, well, not too much), an electric trigger would be easy (simply start the motor)... You could even open the doors (e.g. drive to the designated location, open, wait for the tube to be filled, close).


Hi, I know someone who build a ballast tank for a RC submarine from a syringe and a servo. And when you put it outside the pressure housing, you've the solution for great depths - even if the motor gets wet ... or a more powerful and sealed design inside a housing.

Would it be contaminated when the opening is in the opposite direction or on the side of the movement direction while ascending (negative pressure through movement of the water outside)?


What came also into my mind - why not using such a pump and a plastic bag inside the pressure housing?



This sounds intriguing (the servo/syringe device). Any photos of it online??


Not from the version I saw, it was really "simple". The mechanics from the servo with the motor where directly used. In the gear for the servo arm was a thread? for the threaded rod drilled in at which the plunger from the syringe is attached. With a metal gear servo it should/could be compact and robust.

A more complex version (like a pro series ballast tank) picture: http://www.rc-network.de/forum/attachment.php?attachmentid=847711&d=1341706260


This looks great Eric, I was hoping to make one myself when I got my openrov. It's on order!

When I was working on research boats, the sampling bottles used a piece of silicone tubing for the spring. That way you don't have a corroding piece of metal in your pristine water sample. I think it was also less likely to foul and ruin the bottle seal. The triggers were some kind of solenoid, just a post that would retract and release the two loops of line leading to the stoppers.

I'm interested in seeing more instrumentation and I hope I can contribute useful stuff!


The one I mentioned in the other thread used exactly this...surgical tubing. The trigger mechanism was a sort of reversed mouse trap. The end stoppers were connected to each other with the tubing. On the outside, there was a length of heavy monofilament attached to each stopper. The monofilament was looped at one end. These loops were connected to the trigger mechanism. When the trigger weight was released, it allowed the monofilament loops to be released and the stoppers to close on the sample tube.


Does anyone have Matt Crenshwaw's contact info? I would love to reach out to him to further discuss the "epic trigger" release. I think this device would be useful as I hash out a payload idea for our upcoming expedition.

Any leads that can be provided would be most appreciated.



I believe a setup like the servo's and magnet outside can control endcaps as you want.

No reason to develop again if stuff allready is made..


You could use the same plug Idea (as the original post). Attach a string or wire to the plunger, passed through the syringe hole, with a plug on it. like a little blob of silicone or something like that.


Hi Carsten

you mean something like this?

I made a prototype, quick and dirty.... took me 10mins

can be replaced with my magnetic grabber.

empty and "full" (5ml)




Yes something like that, and if bigger samples needed then get bigger syringething (word?)

Maybe even activate those in the first posts

Best Regards




Have you posted 3d files of your grabber parts and maybe a parts list? If not could you? :D