Electrolysis for dummies

Ya, for the most part. Because reactions are reversible, and 8.3ms isn't long enough for the atoms to spread out.

Some things can still be dissolved. For instance, I'm told 6VAC and hydrochloric acid will "dissolve anything" (including platinum).

Tim

So, (for a water level sensor), do you think that rather than AC, I could use DC on my probes and just "sample" every 5 seconds or so for a very short duration?

A very very short pulse every 5 seconds or so?

Not exactly. Electrolysis still occurs with most metals. The process is as follows:

With silver-plated electrodes and pure distilled water, the positive electrode releases silver ions (Ag+) and the negative electrode releases hydroxyl ions (OH-). The equations are

anode: 2Ag - 2e(-) --> 2Ag(+) cathode: 2H2O + 2e(-) --> H2(g) + 2OH(-)

When the polarity reverses, these ions are now in the vicinity of the electrode, and can combine in a thin layer next to the electrode called the Nernst Diffusion layer. This produces silver hydroxide, AgOH, which is inert and insoluble and precipitates out of solution, or forms a dark black or brown film on the electrode. The equation is

Ag(+) + OH(-) --> AgOH

At the same time, some ions manage to escape into the solution. These move under the influence of the applied field and increase the conductance of the solution. A similar reaction occurs with zinc and copper electrodes, except the copper ion has an ionization state of 2, so twice as many hydroxyl ions are formed at the cathode. The zinc and copper ions do not form hydroxides, but plate out at the cathode as pure metal.

The increasing conductance of the solution shows that free ions are liberated and electrolysis is taking place. This can be prevented by using platinum or carbon electrodes which release oxygen at the anode and hydrogen at the cathode.

The hydroxyl and hydronium ions recombine so there is no net increase in charge carriers, and the conductance remains the same. This feature is used in conductivity meters, such as pure water testers (PWT) or total dissolved solids (TDS) meters. If you use DC instead of AC, ions from contaminants in the water migrate to their respective electrodes and form a cloud around the electrode. This acts similar to the space charge in vacuum tubes and changes the apparent conductance. Changing the distance between the probes changes the conductance in strange and unexpected ways that are not related to the probe separation.

You can avoid this effect by swishing the probes around to disrupt the cloud and minimize this effect. This makes a poor man's conductance meter and is adequate for non-critical tests.

Regards,

Mike Monett

You may run into problems with contamination after a while. Bacteria and other life forms can grow and cover the electrodes.

If you just want to detect water level, why not do the same as a clothes washer? Go to your local appliance store and find where they put the scrap appliances. If you ask nice and leave things neat and clean, they will probably let you retrieve some parts. The water level sensor is a flexible tube going to the drum and connected to a diaphram switch. Most are adjustable, or you can change the water level by changing the amount of tubing immersed in water.

Regards,

Mike Monett

Very impressive Mike...I read it slowly.....you're very fluent on the subject.

I think what I'm going to do is just get a glass of water and breadboard different arrangements and look at the results.

I want to use the input of an optoisolator to sense the water level which will control a pump.....so it's not like the probes will be in the water long....and just a short random pulse would work.....plus the opto input takes very little current to "turn on".

I'll have the output of the opto trigger my PIC microcontroller.

I'm just wondering with such little contact with water whether there will be a problem. (probably)

A problem bigger than electrolysis will in all likelihood be a build-up of minerals on the probes. Each time the probes become wet and then dry out as the level falls any dissolved solids in the water will remain on the probes. Depending on the nature of your water, this may or may not be a major issue, but it is why most modern level control systems do not use conductivity probes, although they were quite popular in the past.

Thanks for the compliment. Yes, there will likely be many problems trying to use conductance. Did you see my other post on using the washing machine solution? You probably couldn't find water more contaminated than in a clothes washer, and the diaphram method can work for decades without failure.

Regards,

Mike Monett

Hi Don,

I did not write the above, as implied in your quoted text, and I do not normally sign my posts with the word "stupidified". I was replying to the OP's question.

Thanks for your url. However, in another post, the OP tells the reason for wanting to learn about electrolysis. It has nothing to do with generating hydrogen. Regards,

Mike Monett

Hi - using the neat XPat feature of XNews, I was able to retrieve the parent and see that you were responding to the OP's post and not mine.

Sorry for the confused response. Anyway, he wants to measure water level, not make hydrogen gas.

Regards,

Mike Monett

Mike Monett a écrit : Yes, there will likely be many problems trying

Your proposed solution of the level-sensing problem with a washing machine sensor is rather failproof. But all the different drawbacks of a conduction based sensor made me think of capacity change. This can be enormous, given the very high value of the dielectric constant of water. And as the original poster seems to want to go rather far, technologically speaking, by using a microcontroller... So long Peter

You just missed the thread on capacitance based level sensors. With these, the electrodes are not in direct contact with the water so electrolysis is not an issue.

Another thing that works well (if the OP has the room for it in his tank) is a float switch. There is a type which is a tilt switch encased in a float that hangs from its cable. The float is buoyant cable end down, so when the fluid reaches it, it tips to one side, closing the circuit.

These are used in pumped sewage systems, which are even more contaminated than most washing machines.

"Paul Hovnanian P.E." wrote: [...]

Don't most sump pumps come with one already installed? They are a bit of a pain to adjust, and I have had problems with reliability. Either not shutting off, or failing to turn on. Sometimes the switch breaks or jams. Sometimes the float sinks. I think these are fine for sewage plants, where someone is constantly available to maintain them. But I have a suspicion the diaphram may be much more reliable.

Of course, you really need two. One to trun the pump off when the level falls below the setting, the other to turn the pump on whe the level rises.

Now comes the tricky part. It needs some form of memory to realize that it has turned on because the water level rose, but it shouldn't run off until it clears the lower sensor. This could be a simple relay, but now we are adding complexity and more failure modes.

So perhaps there is a simpler method, for example using fluid logic, that might be more reliable? Regards,

Mike Monett

Maybe there is. Only needs one switch. This concept is for a sump pump mounted in the basement.

Make sure the pump has an anti-siphon valve at the exit. Most do - it is silly to try to pump water out of the basement and have it return as soon as the pump stops.

Connect the tube for the diaphram switch at the pump housing near the exit and anti-siphon valve.

Adjust the diaphram switch to turn on at the desired water level.

When the pump turns on, the water pressure will keep the switch closed.

When the pump runs out of water, the anti-siphon valve will prevent water from flowing back into the pump. The pressure will drop, opening the diaphram switch. The pump turns off and the system waits for the next cycle.

Of course, there may be some additional details that require solution, or the OP may have a completely different need in mind.

But this eliminates the adjustment problems and unreliability of the float and level switch.

Regards,

Mike Monett

This is the kind of float switch I was thinking about. They are pretty reliable, since they have no external adjustments or moving parts:

The only disadvantage these have is in an environment where they might get tangled up with something.

I've had different problems with sump pumps all my life. Rarely the same thing twice. Anything with wires, rods, levers, floats, plugs that come loose, corroded switches, and so forth. I'd minimize the number of moving parts and keep them protected inside the housing so nothing can get at them.

Regards,

Mike Monett

my chemistry teacher put two stainless steel electrodes in a jar of dilute sodium hydroxide, connected 12V AC (50Hz) and produced a mixture of hydrogen and oxygen gas (Which He trapped in soap bubbles and ignited)

possibly it does some of it

wahtever you do the positive electrodes will dissolve if it contacts the electrolyte. stainless steel is much better than copper

if you want to avoid this you could use insulated probes and measure the capacitance. - earth one and measure the impedance of the other.