Underwater electrode

Hi all, first of all I am really no expert, so I apologize in case I make gross mistakes. I have some ideas in mind for a project and I would like to get some inputs in order to see what's feasible or not and what's the best way to do it.

I would need to design an underwater electrode (no salty water, only fresh) that would produce voltage changes in the order of mV (1-20 mV). It does not really matter if the changes are positive or negative, however it could be nice to be able to control their sign. They definitely would need to be switched ON and OFF by some external trigger but they should also have the chance to stay "ON" for long times (hours or days). They should be shaped in a way that would allow an array arrangement (I was thinking about some sort of "tile"-like flat electrode). The tricky part (or maybe impossible) would be to have the circuit insulated but still able to produce detectable signals from the outside. The water pressure would not be super high... I plan to use them in very shallow water (20cm) on a 80x80cm surface.

I though one way could be to use LEDs as when they light ON they cause a drop in the circuit voltage of some few mV (1-5 I guess). LEDs would have to advantage to be also visible and so if one has an array (as I plan to build) one could identify where the voltage change occurred. the array will be put in a square glass tank and the activity of the electrodes will be monitored through other recording electrodes.

Ultimately the idea is to have an array of electrodes which activity can be patterned and programmed via software (or arduino-like solution).

Any feedback on this or any idea on which materials/electrodes would be best to use in such a configuration ?
As an alternative, would it make sense to use electromagnetic field through the glass tank ? I hope all this doesn't sound too stupid

Thanks a lot !

Hi,

a drawing would be helpful.

Klaus

Rather than describing what you expect from each sensor option, it would be better to specify exactly what you want to measure/detect; For example, it was not clear if you just want to detect the presence of water, or the depth of filling as well. There are factors to consider as electrical insulation, which is a common requirement when it comes to water tanks not totally free from the risk of having any electric potential injected, or that are not grounded. In short, give more details about the end application itself, if possible.

KlausST said:

Hi,

a drawing would be helpful.

Klaus

Click to expand...

Here is a schematic drawing of how it could look like

Thanks Andre, you are right.
I do not need to record. I need the electrodes to produce a constant "detectable" potential that could be set or programmed via PC. The electrodes won´t record. They will just be emitters.

I would record them just in trial sessions while testing (one important issue for example would be that each "tile" of the array, each electrode, would be detected as independent... or at least that there would be enough difference among electrodes so that they could create some sort of gradient). To see whether it would be possible to detect those signals (through amplifiers). The "electrical pattern" would be then used in experiments in which live animals will navigate the "electric field", field that can be changed via software depending on the need. The animal behavior would be videorecorded and that is why I thought perhaps using LEDs could be actually interesting so that one has a visual proof of what the pattern looks like (not only via software eventually).
Hope this clears the doubts. If you have more specific questions of course please feel free to ask.

KlausST said:

Hi,

it´s some research I think.

I d´go for a PCB - or multiple smaller PCBs. Gold plated to avoid corrosion.
On the back side I´d put the electronics. You need to fill the vias, to make the whole PCB waterproof.

Test the circuit without water. When it works you may put some isolating compound on the back side to protect the electronics from water.
Did you consider how to control the inidvidual pads´ voltage?

***
You say animals.. and navigate.
Don´t you think the LED´s light may distort the test results? Maybe the animals navigate rather by light than the low voltage...


Klaus

Click to expand...

Thanks Klaus
I will make a search and check costs and feasibility based on what you suggested. LED would be used with infrared...the animals would probably not detect those. I do not know how to control the voltage but I thought perhaps via PC using some software interface connected to an "external controller"...both things should be designed and I am just trying to collect ideas or suggestions on what is already out there that could be used for this purpose.
What does PCB stays for ? sorry for my ignorance...

Hi,

it´s some research I think.

I d´go for a PCB - or multiple smaller PCBs. Gold plated to avoid corrosion.
On the back side I´d put the electronics. You need to fill the vias, to make the whole PCB waterproof.

Test the circuit without water. When it works you may put some isolating compound on the back side to protect the electronics from water.
Did you consider how to control the inidvidual pads´ voltage?

***
You say animals.. and navigate.
Don´t you think the LED´s light may distort the test results? Maybe the animals navigate rather by light than the low voltage...


Klaus

- - - Updated - - -

Hi,

What does PCB stays for ?

Click to expand...

with your project I assume there will be a lot new terms.

I recommend to do an internet search, google, wikipedia..... then you get faster and more detailed response

Klaus

As an alternative, would it make sense to use electromagnetic field through the glass tank?

Click to expand...

Static electrical field doesn't propagate through glass wall (or any isolating electrode surface), it's shorted by a conductive liquid.

Immersed bare metal electrodes can generate a static field with respective current flow according to liquid conductivity. Unfortunately the field distribution can be modified by a double layer.

Presuming that you don't want the electrode to corrode or
taint the solution, I recommend getting some welding
"gouging rods" which are graphite with a copper jacket.
Peel off the copper where the rod will be submerged but
leave it where it's dry, for a nice solderable connection.
I have used these for homebrew electrolysis and plating
experiments, the graphite is nicely inert.

Have you tried experimenting with voltages generated between two different metals when immersed in electrolyte? Pond water conducts electricity slightly better than distilled water, maybe well enough so that you'll measure a fraction of a volt. It's not as complex as your array, but it may influence underwater creatures and yield interesting observations.
They may be injured by exposure to certain metals such as lead and cadmium.

BradtheRad said:

Have you tried experimenting with voltages generated between two different metals when immersed in electrolyte? Pond water conducts electricity slightly better than distilled water, maybe well enough so that you'll measure a fraction of a volt. It's not as complex as your array, but it may influence underwater creatures and yield interesting observations.
They may be injured by exposure to certain metals such as lead and cadmium.

Click to expand...

@Brad: Actually no but the presence of metals should be very limited if not avoided. In the case of PCB obviously there is metal but is probably a thin layer so it could be OK. In the case of LEDs there is metal too but they are tiny... In addition the pattern should be re-arranged on demand so in case of simply displacing object this would take time and would not be precise.

@Dick: I am not sure if I got it right but I thought one good way to avoid "classic metal conductors" could be to use graphite or carbon coated electrode - for similar reasons you mention. Is there a way to have them on a PCB ? perhaps one that comes already like that... it shouldn't be an issue to find it, right ?
The gold plating suggested by Klaus could also be good but still... the material choice would be important also to not disturb the animals in the tank (contrarily to me, they don't like metal).

In the case of LEDs there is metal too but they are tiny...

Click to expand...

LED terminals must not be exposed to the water, otherwise the electrical field will be severely disturbed because LED voltage is much higher than the discussed mV electrode potential. It's essentially that the LED terminals are sealed, a resistance measurement against electrodes should give high MOhm numbers.

As for the electrode surface finish, I agree that suggested gold plating is a good option. The standard PCB flash gold is only a few µm, I guess it has limited lifetime if permanently immersed in water. Thicker and more expensive "contact" gold plating may be optionally used. If you want a metal free surface, carbon print should be checked. But I have no idea if it's long term stable in water.

In the case of PCB obviously there is metal but is probably a thin layer so it could be OK.

Click to expand...

Error of reasoning, I think. It's only the surface that interacts with the liquid and possibly "poisons" it by dissolved metal ions. No matter how thick it is.

Just a query for anyone that could advise me on this:

What would you suggest to easily interface a small PC to a laptop (via serial port or USB) in order to test materials and conductivity? At this point I would try to avoid any arduino like concept, if possible.

Thanks in advance!
Livio

Livio said:

interface a small PC to a laptop

Click to expand...

Does this refer to two computers? Running Windows? Networking by wifi has made it easier in recent years.
Or by ethernet cable. In some cases you may need a 'crossover' cable, especially if the computers are older.

Hi,

Just a query for anyone that could advise me on this:

What would you suggest to easily interface a small PC to a laptop (via serial port or USB) in order to test materials and conductivity? At this point I would try to avoid any arduino like concept, if possible.

Thanks in advance!
Livio

Click to expand...

Since I saw the picture of your electrode I asked myself how to drive it.
In the picture it is an array of 12 x 20 = 240
I don't know if this count meets the real circuit...
But this calles for a multiplexed excitation.
And a multiplexed excitation calls for a microcontroller based system.

I wonder if one can misuse a (LED) graphics controller.

What range and resolution in electrode voltage do you need?
How fast do you want to change it?

Klaus

Thanks Klaus. In theory it could be as fast as LED and the change should be a "translation" so that the pattern of activity would move origin, locked to an object moving above the array. I thought to use OpenCV to "feed" the coordinates of the "origin" to the board whenever a movement is detected.

However this is not the main point now, I will think of it later. Now I need to find a good way to make a simple trial.

For the design of the PCB I was thinking about a "concentric" arrangement (not like in the figure). But maybe this could be more complicated and also decided later. The trial I would like to make now would be using a small array (3x3 or 4x4) as I said just to test the material and the detectability of the signal underwater. I agree with you about multiplexing and a controller system ... Now I wonder if I can do all with a PC/laptop, a small PCB and few other components. just for trial purposes....
otherwise my feeling is that I would be investing a lot of time into something that doesn´t even meet the basic requirements: detectable and distinguisheable signals and a stable array.

FvM said:

Static electrical field doesn't propagate through glass wall (or any isolating electrode surface)...

Click to expand...

Glass is insulating and static electric field will have no problem passing through a layer of glass. Conducting metals are problematic because the surface must be equipotential. However, liquid water is only weakly conducting and can allow static electric field to a degree. The story changes if we consider RF electric fields (low frequency RF can go considerable distance in water).

A metal immersed in a liquid (solution) produces a potential with respect to the liquid (Helmholtz double layer potential) but it is rather messy. In the present case, the results will be meaningless / random noise.

Hi,

PC/Laptop and software will not be a problem.

****
You need the mechanical and physical solution first.
* Base material (Maybe PCB material FR4 = glass filled epoxy)
* conductive surface (Cu with Au finish, or carbon)
* geometry/shape and arrangement of the electrode pads: (square, round, hexagonal like the cells of a honeycomb. In line or interleaved. Mind to have equal and wide clearance to the next electrode to avoide electro-corrosion at the edges.

Then you need to solve the electrical problem:
* how to conduct the individual electrodes
* how to generate the electrode voltages
* how to interface to the PC
* avoid noise and mains hum at the electrodes with respect to the water`s electrical potential. Consider to use galvanic isolation.
(Maybe a battery operated system with bluetooth or IrDa as interface could be a wireless solution)

Klaus

- - - Updated - - -

Hi,

Glass is insulating and static electric field will have no problem passing through a layer of glass.

Click to expand...

Really?
This means you can measure the tiny mV of the electrodes on the opposite side of the glass?

Klaus

This means you can measure the tiny mV of the electrodes on the opposite side of the glass?

Click to expand...

Measuring voltage should be no problem: you will need an electrometer. Larger voltages could be measured with an electroscope. Measuring current many be tricky.

Static electric field goes in a vacuum: if the electric field is constant the voltage difference will be zero between any two points.

Am I missing something?

one PC. it´s not important if ethernet or USB

do you have suggestions for the software and hardware components ?