Building a small scale KCLO3/04 Cell

[MrB]

Your cell designs will have to be very specific if your going to be able to work with 13.8v & 7A. So the second one is no good. The first one... Well, no idea on the quality, and 10A max might be a bit on the weak side, but again, depends on the cell design. The size of the anode/cathode, and the wanted current density, will decide if that is good enough.

[insutama]

i think im gonna try my laptop power supply first because i already have it and dont have to drop more money on a expensive power supply or try and build the ATX one. i want simplicity for this. my laptop power supply is 19V 4.74 A

[MrB]

Referring back to my previous statement. Your average cell will want 5-7v, and high A. Getting something to run with a high voltage, low A, will be a challenge, requiring very specific design-choices, to the extent where you, using the laptop PSU, will have to run something along the lines of 4 cells with tiny anodes.

 

What i'm really saying. It's your life. Wasting time and effort on something that isn't worthwhile is an option you have. Call it exploring, science, or just plain fun, and you have a reason for doing so, and it's all fine. If your purpose is a working cell, then this isn't the right way to get there.

[insutama]

okay so what would you recommend for my small cell id rather not have to build one.

 

If i used a high voltage and low current in one electrode what would happen ?

Edited July 10, 2017 by insutama

[Baldor]

As stated before, nothing beats an ATX power supply in the bang for the buck department, you only need a little tweaking. They are made in such cuantities that the economies of scale drive the price down. Also, are pretty reliable.

 

The first one is a nice lab power supply to have if you are in the electronics hobby, but if not, you are paying a lot for using it at 5V and 10A max. From a 500W ATX It's possible to draw about 20A if you use all the 5V cables at the same time (Look at the label of the power supply) Take a look at this, is the first one I found the specifications online. 20A at 5V for a 450W supply. I will recommend to go big, and get a 1000W ATX, Never hurts to have to much current available, and who knows what your cell will draw, or your future cells.

[insutama]

Good point ill look for a 1000 watt ATX power supply is there any you recommend ?

[insutama]

I can get a 500w atx power supply for 50$ and a 1000w for 240$ so i may cheap out i could get 2 500 w supplys for cheaper lol

[Baldor]

I can get a 500w atx power supply for 50$ and a 1000w for 240$ so i may cheap out i could get 2 500 w supplys for cheaper lol

Search harder.... https://www.ebay.com/sch/i.html?_from=R40&_trksid=p2380057.m570.l1311.R1.TR2.TRC1.A0.H0.X1000w+atx.TRS0&_nkw=1000w+atx+power+supply&_sacat=0

 

Not a good idea to parallel two power supplies. It can be done, but it's asking for troubles and smoke. And we want our smoke out of our fireworks, not out of our tools.

 

You can get about 30A in the 5V rail, or up to 60A from the 12V, but then you will need a beefy and expensive step down converter to get the desired 5V.

[MrB]

If i used a high voltage and low current in one electrode what would happen ?

 

You would kill it, in a single run.

 

I can get a 500w atx power supply for 50$ and a 1000w for 240$ so i may cheap out i could get 2 500 w supplys for cheaper lol

 

Running two PSU's would be like running multiple 12v rails. You don't really want to combine them, without diode bridges making sure they don't feedback on one and other, and let out the magic blue smoke.

[Baldor]

Also, you don't ned a new supply. Maybe you can find an old PC for less than a new supply, and just gut it. It's what I did with this:

https://goo.gl/photos/cu9rh5rL3et921iW6

Curiously, it's an old 300W one, and the 3.3V and 5V rails are both beefier than the 12V one. (20A, 25A and 15A respectively) One old one could be better for what you need than a new one, as long as you can read the label before getting it.

[Baldor]

MrB, what is the ideal voltage for this reaction? Maybe a better solution could be found.

[MrB]

Curiously, it's an old 300W one, and the 3.3V and 5V rails are both beefier than the 12V one. (20A, 25A and 15A respectively) One old one could be better for what you need than a new one, as long as you can read the label before getting it.

 

That is to be expected. Older systems used the 5v rail for powering CPU and memory, today they use the 12v rails.

 

MrB, what is the ideal voltage for this reaction? Maybe a better solution could be found.

 

There really isn't one. It depends on everything from what you are trying to produce, chlorate or perchlorate, sodium, or potassium, with what anode, the distance between the anode and cathode, and the current density you want to run. That is why an adjustable PSU is much more "user friendly" then a static 5v setup. To get the "right" numbers with a set voltage, you have to play around with the spacing until you get all the numbers to line up, with an adjustable you get the spacing "close enough" and do the rest on the PSU.

 

The best "general" figures around say 3-5v for chlorate production, and 5-7v for perchlorate.

[Baldor]

So, and I'm going on a limb here... . Imagine you connect two cells in series, to a 12V supply. Ideally, you will have a voltage drop of 6V at every cell, ideal for perclorate. If you use three cells, you will have 4V, just in the middle for clorate. Of course, the cells will be slightly different, so, no 100% true. This could simplify the search for an adequate power supply, since it's easier to find high current 12V power supplies than 5V, at the cost of using more electrodes.

 

Am I very, very mistaken?

[MrB]

To some extent that works. But it's going to be a pain to maintain. If spacing between the anode and cathode isn't exactly the same between both, or all three cells, your going to have one sell that is faster, and, or, slower then the other(s). When the faster one starts to deplete the supply of KCl, the current goes down, for all the cells, and so on. People do it, it's quite possible, it is just not optimal.

And, 4 & 6v isn't "optimal", for a given product, as i said, it comes down to the distance between the anode & cathode. You might find that a given cell needs 5v for it's expected, or desired current density.

[insutama]

I just bought a 24 pin connector terminal board so now im have limited myself to a 24 pin power supply.

So should i be looking for a older ATX power supply seeing how there is more amperage on the 5v rail ?

[insutama]

http://www.ebay.com/itm/KENTEK-650-Watt-12cm-Fan-ATX-600W-650W-Black-SATA-PCIE-Power-Supply-Quiet-Gaming/221217505749?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D555018%26algo%3DPL.SIM%26ao%3D2%26asc%3D41375%26meid%3D012890210e924dd6a377ca7c73b762ca%26pid%3D100005%26rk%3D4%26rkt%3D6%26mehot%3Dpp%26sd%3D332292693692

 

hows this one ?

[Baldor]

Seems adequate.

 

For better performance in this application, look at what 5V wires are really connected to the board cut all 5V wires from the ATX conector less one of the conected, cut all the ones in the other conectors, and wire them together. The same for ground.

 

Then, the board will control on/off, and you can connect low current draw accessories like a fan or a voltage/current meter to it

[insutama]

How can i do that before i buy it tho

[Baldor]

You already have the board, I'm right? You don't need to do anything before buying the power supply, only check the board.

 

Just check in the board what pins are connected to the 5V output, to ground and to 3.3V output. Use this as reference. You can use a multimeter to check, or you can check visually what pins have traces connected to them. You only need one 5V, one 3.3V, one 12V and one ground to the board, plus all the control wires.

 

For example, you can check pins 4,6,21,22 and 23 to see if the are connected to the 5V outputs. Imagine only 21 and 22 are connected to the 5V output. When you receive the power supply, you can cut 4,6, and 23, and group them with the other red wires from the other connectors, to have a connection capable of sourcing a lot of current. You can also cut 21 or 22, one wire will be enough for the small things you will connect to the board.

[MrB]

If he bought the one you linked earlier, 21,22, & 23 is connected for 5v, all the others just have a single pin.

[insutama]

actually this is the one i bought http://www.ebay.ca/itm/192105522335?_trksid=p2060353.m570.l6004&_trkparms=gh1g%3DI192105522335.N41.S1.R2.TR3

[insutama]

So i should be looking for a power supply with high amperage on the 5V correct ?

[MrB]

actually this is the one i bought http://www.ebay.ca/itm/192105522335?_trksid=p2060353.m570.l6004&_trkparms=gh1g%3DI192105522335.N41.S1.R2.TR3

 

In that case... All of the 5v pins in the 24-pin connector, is connected to the 5v terminal. 3.3, 12, and -12v are all on a single pin.

 

So i should be looking for a power supply with high amperage on the 5V correct ?

 

If you are looking to run on 5v, yes. If you are looking for "used" parts, try looking in PC cases that say 'AMD Athlon 64 X2' "or similar" stuff, since they were running of 5v much longer then Intel stuff did, and hence the most power used on a 5v rail, would be a high power AMD system.

Intel started to shift to 12v CPU powerphases during the PIII era. Some late OEM motherboards have the 12v ATX CPU connector, and by the time the Northwood P4 came around, pretty much all Intel motherboards were running 12v for the CPU power phases. It should be noted that "higher end" AMD gear switched to 12v CPU powerphases at the same time as Intel did. Lowend stuff refrained from using it for much longer, so, OEM systems could have pure gold, in the 5v sense.

 

As far as 12vs 5v CPU powerphases... One really isn't "better" then the other. It's not an AMD vs Intel thing. To power a high end overclocked CPU, you could need 150w of power. That is 30A@5v, or 12.5@12v. It's simply a matter of how much can you draw over the copper traces in the motherboard, and the wire feeding the power to the board. (Which is why the 12v dedicated feed always is damn close to the CPU on the motherboard. Short traces, doesn't use as much area. Cheaper, and easier construction.) But the chips needed to step down to the voltage the CPU wants, is a bit more expensive when you work with 12v as compared to 5v. Anyway, oftopic and such. Look for late AMD Athlon, preferably 64 X2 systems, and a dead giveaway when you look in to an old PC... If there isn't a 4 or 8 pin connector with 2-2 or 4-4 black / yellow configuration, then the PSU is older then ATX12v1.x and should have the majority of it's watt's available over the 5v rail.

If you find a even older PC powersupply, then it wont have the ATX powerconnector, and will have pretty much all it's power available on the 5v rail, but an A PSU is probably over 20 years old by now, which might be.... well, bad.

 

Come to think of it. ATX 12v1.0 came around with the millennium shift, so it's freaking old to. That old PSU's might have issues, try and stay safe when you test them.

Good luck.

[insutama]

 

In that case... All of the 5v pins in the 24-pin connector, is connected to the 5v terminal. 3.3, 12, and -12v are all on a single pin.

 

 

 

Is that what i want ?

 

also i dont understand how the 12v would be even of use to me for making chlorate isnt optimum voltage 5 ?

[MrB]

Is that what i want ?

 

Who knows. I wouldn't actually run the cell of one of these, but if you are, then 5 connected pins is better then 1.

 

also i dont understand how the 12v would be even of use to me for making chlorate isnt optimum voltage 5 ?

 

As i said before. A lot of PSU's are designed in a way that forces you to put a load on the 12v rail, to get a stable 5v rail. You wont know until you try, and again, as i said before, i'd test it out with a dummy load consisting of 12v lightbulbs on the 5v rail, and measure my voltages, just to make sure i don't damage an expensive anode.