Sciencemadness Discussion Board - General questions on chlorate cells and building a controller

General questions on chlorate cells and building a controller

Little_Ghost_again - 6-10-2014 at 14:01

Hi all

I have UTFSE, and read a good few posts but I have some questions. I am trying to make a controller for cells (micro controlled).
One thing I noticed was in a post it mentioned that when the power goes off the electrodes can be damaged?
How much current passing into the cell would it take to protect a cell that had lost power?
What I am thinking is a battery back up pushing just enough mA to stop damage but not actually run the cell from, that way if it went offline over night it could switch to battery backup and save the electrodes from damage.

The other question I have at the moment is for temperature sensing, I am looking at a way to enclose a sensor into a glass tube, I am guessing mineral oil inside this would be ok? I dont want one to break and the oil react with the cell contents causing a problem.

Most the posts seem to indicate cells run at 5V or some at 12v, would PWM be ok in a cell? or would it be better to PWM a voltage controller and then buffer that so the voltage was steady? I was thinking a PWM of around 15 kHz.

I am also looking at a way to adjust the Anode and cathode distance by small amounts via servo control, but cant find a good way to seal the thread I would need.

jock88 - 6-10-2014 at 14:07

You are getting very complicated.

There is some reading on Chlorate making here.

http://oxidizing.typhoonguitars.com/

Use a constant dc current supply with an open circuit voltage of about 6 volts. No need to move electrodes around.
Don't bother monitoring temp. just make sure it does not get too high.
Set up a device to drop acid into cell to control pH as this will dramatically increase current efficiency (time efficiency).
Put a 3 or 4 volt rechargable battery across cell if you are worried about power cuts just to keep some current going into cell (there is hardly any need)

People run there cells at 5 or 12 volts as they are using salvaged computer supplies and are <B>STUCK</B> with a constant voltage power supply that puts out 5 or 12 Volts.

[Edited on 6-10-2014 by jock88]

Metacelsus - 6-10-2014 at 14:09

What about the cell are you trying to regulate?

Use 5 volts; 12 will corrode your electrodes pretty quickly. I've never heard of turning the cell off ruining the electrodes.

Could you melt wax inside of the tube instead of mineral oil? That way, if it breaks it won't mix with the cell.

Little_Ghost_again - 6-10-2014 at 14:26

Yes wax is a better choice for me! Thats a great idea.
I know it seems over complicated but the idea is more for a electronics project, I like the idea of a automated cell thats monitored and logs the data. I will use a pc psu but varying the voltage is easy enough, either by adjusting the pwm in the psu so the output voltage is different or building a voltage regulator.
My own cell I will keep below 5A, but the controller will be able to handle a lot more.
I have decided to dump all the data to sd card then it can dump that when connected to pc via serial.
The pH monitoring I will need to sort, I am looking for way without using a pH probe (maybe gas production or current shift?). the temp control was part of the moving electrodes thing. But I am open to ideas

thanks alot for the ideas

Little_Ghost_again - 6-10-2014 at 14:31

Whoa thats a cool site in the link. I will hold off until I have had a chance to read through it all

Artemus Gordon - 6-10-2014 at 14:50

Quote: Originally posted by Little_Ghost_again

Hi all

I have UTFSE, and read a good few posts but I have some questions. I am trying to make a controller for cells (micro controlled).
One thing I noticed was in a post it mentioned that when the power goes off the electrodes can be damaged?
How much current passing into the cell would it take to protect a cell that had lost power?
What I am thinking is a battery back up pushing just enough mA to stop damage but not actually run the cell from, that way if it went offline over night it could switch to battery backup and save the electrodes from damage.

The other question I have at the moment is for temperature sensing, I am looking at a way to enclose a sensor into a glass tube, I am guessing mineral oil inside this would be ok? I dont want one to break and the oil react with the cell contents causing a problem.

Most the posts seem to indicate cells run at 5V or some at 12v, would PWM be ok in a cell? or would it be better to PWM a voltage controller and then buffer that so the voltage was steady? I was thinking a PWM of around 15 kHz.

I am also looking at a way to adjust the Anode and cathode distance by small amounts via servo control, but cant find a good way to seal the thread I would need.

Does PWM mean pulse-width modulation? If so, you don't want that in the reaction vessel. You want a DC voltage as smooth as possible. The voltage needs to be a few volts higher than the textbook voltage of the reaction you are trying to induce in order to overcome resistance in the circuit, but too much voltage will just turn into wasted heat. You should calculate what level of current you want and then control your power source to maintain that current level. But if your wires get too hot to hold comfortably you need to back it off, or get fatter wires.

As for a heat sensor for the reaction vessel, what will your reaction vessel be made out of? If it is thin glass, or some other fairly good heat conductor, I would just tape the sensor to the outside wall. I don't think you need to be more accurate than maybe 10 degC, because what you are trying to do is prevent boiling and to avoid energy waste due to excess voltage. In fact, if you make sure your current holds steady at your chosen level, a temp. sensor is really gilding the lily.

I see no reason to worry about power failure causing damage to your electrodes. I assume you will be using sufficiently inert electrode materials. Once again, over-voltage would be a more likely cause of electrode damage.

Electrode spacing, I would think is not super-critical. You want them fairly close to reduce resistance to current flow, but letting them touch would be bad. Most dedicated electrolysis apparatus I've seen keep 1-2 inches between the electrodes. BTW, do you plan to have a fuse or other circuit-breaker? THAT is much more important than automatic temperature control.

jock88 - 7-10-2014 at 04:27

Quote: Originally posted by Little_Ghost_again

Yes ...I will use a pc psu but varying the voltage is easy enough, either by adjusting the pwm in the psu so the output voltage is different or building a voltage regulator.
My own cell I will keep below 5A, but the controller will be able to handle a lot more.
I have decided to dump all the data to sd card then it can dump that when connected to pc via serial.
The pH monitoring I will need to sort, I am looking for way
thanks alot for the ideas

Build a current regulator, not a voltage regulator.

The pH is easy to monitor (there is no need to if you are just making Chlorate) but you need an expensive pH probe. The cheap ones will fail in a short time.

Little_Ghost_again - 7-10-2014 at 04:39

Looks like I have got a few things wrong!! That site is really good I have a rough idea what I am aiming at now. I will build the current controller but my PSU has so much ripple I think the electrolytic caps are on the way out, I might end up using a different transformer and building the PSU from scratch.
The ATX psu looks awful on the OScope! 750mV of really bad ripple around mains frequency. Strange I was expecting it to be at a higher frequency than that. I guess no harm changing the caps see if there is any difference, if not I could addd some smoothing and regulators.
I have some ten turn pots, total overkill but would give very fine current control when used with a regulator and bypass MOSFET

[Edited on 7-10-2014 by Little_Ghost_again]

phlogiston - 7-10-2014 at 06:38

That site was originally written by the member 'dann2'. I see his last post was from jul 2012, so he is currently probably not active here .

I have always been wandering if you could perhaps use the conductivity of the cell liquor to establish the endpoint. Or better yet, use amperometry to determine its composition during the run. Never got around to investigate it, but possibly fun to experiment with in the context of your project.

Little_Ghost_again - 7-10-2014 at 08:22

Quote: Originally posted by phlogiston

That site was originally written by the member 'dann2'. I see his last post was from jul 2012, so he is currently probably not active here .

I have always been wandering if you could perhaps use the conductivity of the cell liquor to establish the endpoint. Or better yet, use amperometry to determine its composition during the run. Never got around to investigate it, but possibly fun to experiment with in the context of your project.

Thats kind of why I want to log all the parameters, or as many as I can. It seems logical that voltage/current will change in some way according to what is going on in the cell, same with gas production.
The micro I am using can measure temperature to 12 bits so 0.01c, the ADC voltage/current measure can read down to 1mV/1mA (with some didgery doo dah ing.
Gas production I can figure out and if my idea works then I can measure it down to 0.01ml per second (burette and capacitive sensing of oil with servo off gassing).
The gas thing I have kind of done before for my fish tank (the plant one) with CO2, it wasnt as sensitive but I didnt have a burette then. Or I can try IR sensing in the burette.
But the idea is too measure everything I can and have it logged every second

hyfalcon - 7-10-2014 at 10:30

Two .75" X 10" strips of MMO on the way.

Little_Ghost_again - 7-10-2014 at 11:07

Quote: Originally posted by hyfalcon

Two .75" X 10" strips of MMO on the way.

Thank you very much!! I am looking about for some titanium to attach them, I might be able to get some at school

hyfalcon - 7-10-2014 at 11:12

These are long enough that they should stick up out of your reaction vessel. Just clamp directly onto the expanded titanium mesh for your electrical connections. Maybe cut a couple of slits in a piece of plastic to protect your electrical connections.

phlogiston - 8-10-2014 at 01:22

It would be best to use non-combustible materials. Oil presents a hazard in chlorate cells, and may also not be the best material to use in a gas production meter if the gas may contain a high % of oxygen or chlorine (or maybe even ozone under certain conditions).

Little_Ghost_again - 8-10-2014 at 01:51

Quote: Originally posted by hyfalcon

These are long enough that they should stick up out of your reaction vessel. Just clamp directly onto the expanded titanium mesh for your electrical connections. Maybe cut a couple of slits in a piece of plastic to protect your electrical connections.

Thanks I wasnt looking forward to trying to tap some titanium

.

I am not going to use oil now I will use a low melt wax encased in glass, I have some pipettes with a fairly large glass bulb on, so will put sensor in a cut one and seal one end, the other end (pretty long) will poke out the top and have the cable sticking out, I can seal it with silicone on cable end and heat seal the other.
I will start making a bubble counter later, basically a coil of tubing that slows and separates the gas bubbles, these pass over a sealed area with a photo transistor and IR emitter, this should give accurate counts then all I need to do is calibrate by measuring how much gas is in say bubbles (burrete displacement).

Now yes I know this looks totally mental for a chlorate cell, but for me this is a chance to do a cool electronics project AND learn exactly what happens inside a cell. It might even provide some good data that helps others.
I have also decided to try and weigh the cell continuously, this is way harder for me, I dont have a sensitive pressure sensor, but I do have some very cheap pocket scales that way 300G in 0.01 resolution (so 10mg), they have a min weight of 50mg but thats not a problem.
What I could do is take them apart and see if I can tap into the chip or maybe read the sensor with a micro.
if I can get the weight as well every second then I can build a complete picture.
I might use a separate way to measure the TDS by conductivity, no idea how useful it will be until i try, I intend to just pulse a fixed voltage (3.3V) for 100ms every 5 seconds, I can then measure the current from this and get an idea of the Conductivity, I will switch over the polarity each cycle to spread the wear.
I have access to matlab so the idea is to dump all data into that once a day or on command and have it do all the working out and graphing.
I am looking for a very cheap pump for top up, I might use a old windscreen washer pump and bottle (one in garage)..

hyfalcon - 8-10-2014 at 03:21

One thing you are going to notice if you go for making potassium chlorate directly from potassium chloride is solubility issues. Any time chlorate levels build up, due to it's solubility, you will have crystals everywhere clogging anything and everything. When you add di-chromate to the mix things get interesting if you try for an active circulation. When that liquor hits cooler temps it looks like a snow storm in the reaction vessel. That's why I use glass on the smaller cell I run.

Little_Ghost_again - 8-10-2014 at 03:41

Quote: Originally posted by hyfalcon

One thing you are going to notice if you go for making potassium chlorate directly from potassium chloride is solubility issues. Any time chlorate levels build up, due to it's solubility, you will have crystals everywhere clogging anything and everything. When you add di-chromate to the mix things get interesting if you try for an active circulation. When that liquor hits cooler temps it looks like a snow storm in the reaction vessel. That's why I use glass on the smaller cell I run.

I am looking for a glass vessel, I have a schott bottle but its a bit small, I have some 1 Ltr wide necked conical flask I could use and just make a cap for it.

Define "When you add di-chromate to the mix things get interesting"

Interesting I like but that sounds more like take cover

I am aiming at sodium to start with and will take it from there. I could cut the mesh and run a couple of smaller cells and compare results. Nothing is in stone yet.
Just tweaking my circuit diagram and taking the scales apart to see if I can read the sensor. Then I can design and etch the circuit board.
I have swapped chips again and gone for a 32 bit energy micro wonder gecko (real over the top overkill), but it will give me very fast reading and high resolution for the ADC.
decide for the test run I am using 12V regulated down to around 4.5V and current regulation upto 5A though i will probably try and run it around 2-3 A.
Still reading what I can get my hands on

hyfalcon - 8-10-2014 at 08:37

Quote: Originally posted by Little_Ghost_again

Interesting I like but that sounds more like take cover

The reading never ends, at least not for me anyway. Always something else to pick my interest. Just be cautious when using di-chromates. Turn your power off and let the mist settle before opening your reaction vessel. You don't want atomized di-chromate mist in the air ready to be breathed in. ALWAYS wear safety equipment. Gloves, at least long sleeves and pants, lab coat preferably, full face shield for the splatters when collecting chlorate. You really want to avoid di-chromate on the skin.

[Edited on 8-10-2014 by hyfalcon]

jock88 - 8-10-2014 at 14:26

There is no point in the amateur using dichromates in a chlorate cell to increase current efficiency by some useful amount. They are not worth the hassle.
They work by putting on film on the cathode which effectively raised the current density on the cathode. It is much easier to just use small cathodes and forget about additives of any kind.
Keep chlorate making family friendly and as green as possible

hyfalcon - 8-10-2014 at 14:56

He's going to have 2 DSA electrodes to work with so I guess it won't matter. I've always used them myself, but can't tell it by the third recrystallization anyway. I love watching a saturated solution of KClO3 approaching the freezing point and the crystals starting to fly.

Little_Ghost_again - 9-10-2014 at 06:18

Quote: Originally posted by jock88

There is no point in the amateur using dichromates in a chlorate cell to increase current efficiency by some useful amount. They are not worth the hassle.
They work by putting on film on the cathode which effectively raised the current density on the cathode. It is much easier to just use small cathodes and forget about additives of any kind.
Keep chlorate making family friendly and as green as possible

Ok as I dont know one way or the other I will do both, a cell with dichromate added and a cell without. With all the logging of data any differences should stick out.