Electrolytic Cell Power Supply

[mrjeffy321]

I am working on trying to get a reliable power supply primarily for my Electrolytic Cell, and secondly for any other power supply needs that should come up.

I have pretty much concluded that a computer power supply will suit me best. In the past, I hav been using a AT power supply, but I never got it to work very well, it generally would output very low and I was forced to use only the +5V output or else it would shut itself off.
I am now experimenting with ATX power supplies. I have two I am working with trying to see which one is the better choice.

In my electrolytic cell, the more current I can get the better, because this will reduce the time I need to run the cell (the difference between months and weeks of continuous opperation). I took some readings using both power supplies in similar conditions to what would be used in real usage.
For the first power supply:
+5V
-Volts = 4.9
-Amps = 1.10
-Calculated Ohms = 4.45

+12V
-Volts = 11.1
-Amps = 4.6
-Calculated Ohms = 2.4

For the second power supply:
+5V
-Volts = 4.9
-Amps = 1.21
-Calculated Ohms = 4.05

+12V
-Volts = 10.6
-Amps = 4.5
-Calculated Ohms = 2.35

the two were similar to each other. There were however a few things I noticed. First, the calculated resistance using tghe same cell, is remarkably difference between the +5 and +12 V outputs on both power supplies, I can only assume that there is some type of built in resistance. Also, on all the reading I took (except for 1, I'll get it it in a second), the current was slowly rising, asd I watched the meter, it would gradually increase, which I would suppose means that the resistance though the cell is changing (getting smaller) the longer the cell opperated. This is great that I am getting a high current through the cell, but it won't take long for the current to over load the power supply and it will shut it self down (assuming there isn't some minimum resistance the cell will eventually get to, but the question is whether it will over load the power supply before it gets there or not). There was one output, the +5V on power supply 2 that was pretty steady. it held very close to 1.2 amps for as long as I sat there and watched it, and then some, occassionally it would drop down to about 1.19 or get up to 1.22, but compared to the others, it was constant.
Is there something in the cicruit that will limit [I can't believe I am asking this, I normally would want the max current I can get], the current so that it will keep the power supply from over loading and shutting itself off if the amps should get too high, it will simply increase its resistance automatically to maintain the current? I doubt it is that simple.

Any other suggestions on how to make this power supply work idealy for my application?

 

[Cliff_J]

If you add a resistor in series with the load it will drop voltage seen by the load and thus limit the current.

Something like an automotive light bulb would have the ability to take advantage of a self limiting feature where as more current flows and it heats up its resistance goes up and thus it eventually reaches an equilibrium point for how much current it will allow to flow. Plus it has this handy glowing effect so you get some visual feedback on how much current is flowing and there are a lot of different sizes for the bulbs.

It could be your meter that's giving you some of the changing readings, after using a cheapie meter a couple weeks ago I am again reminded why I have a Fluke 87 in my tool kit, and even though its a long ways from being lab grade its still plenty good enough for most tasks.

 

[willib]

if you google atx power supply
you will find explanations on how to wire them to give you full power..
there is a connection that you have to make , can't remember though ,to make it output many more amps than you are getting..

 

[mrjeffy321]

if I add a resisor into the circuit, won't that just reduce the total current I can get, not necesarily make it constant?
Can I expect that much of a change in a light bulbs internal resistance as it gets hotter and hotter? How hot are we talking about, I could just as easily put in a real resistor to do the same job couldn't I?
Maybe it is my meter, but I doubt it, the current was rising about .1 or .2 amps every minute.

I have already seen a lot of those pages before on how to make an ATX power supply output its full power, like this,
http://web2.murraystate.edu/andy.batts/ps/PowerSupply.htm
the main thing is that it needs a minimum load on its +5V output (and sometimes +12V too), to make it think that it is working the way it should be. Mine appearently done need that, it must be build in somewhere already, because the work the same with or without an extra load in its outputs, all I need to be sure to do is connect a certain (unlabeled) to ground and it will then turn on and output the correct voltage.

 

[willib]

Thats exactly the page i was thinking of..

In my electrolytic cell, the more current I can get the better, because this will reduce the time I need to run the cell (the difference between months and weeks of continuous opperation).

please explain , is this an experiment , your electrolytic cell ??
EDIT::nevermind i just saw you explanation in th other thread..

 

[mrjeffy321]

The higher the current, the higher the charge per second is flowing into my cell. The charges that are flowing are electrons, and I am using those electrons in a chemical reaction to turn NaCl into NaClO3 (with some other stuff). So the more electrons I can put into the reaction in the shortest amount of time would mean that I would have to wait less for it to complete.
To give you an idea of how long this is.
If I were to have 1 liter of water with 300 grams of sodium chloride dissolved in it, it would need 1100 amp hours to convert 90% of the NaCl to NaClO3 assuming 50% efficency. If I ran the cell at my old average current of 1.5 amps, it would take almost 31 days of non-stop opperation to complete.
So that is why I want the highest current I can get, I don't want to wait a whole month. If I stepped up the current from 1.5 amps to 5 amps, then I would have to wait only 9 days, see the difference.

There are 2 ways I could go about increasing the current. I could raise the potential differnce between the two electrodes and keep the resistance the same, or I could keep the potential difference the same, but lower the resistance, or both.
I can't do much about the potential difference, since I want to use a ATX power supply which outputs a pretty fixed voltage but I can however mess with the resistance to an extent.
To lower the resistance, I could ...
-reduce the spacing between the electrodes, the closer they are, the less resistance. But this is a tricky game to play. I don't want them too close, I want there to be a good flow of water around them so to more efficently electrolysize the entire water, and not just a select local area. I don't want them too far apart as to make the resistance too high.
I could also toy with the concentration of electrolyte in solution. more electrolyte, the less resistance (up to its maximum solubility), but I can't mess with this because of the precise calculations needed to determine the time, yield, and other things involved with the reaction. If I were to "willy nilly" add more salt in the middle of the reaction, then I will have just that much more to electrolysize (or left it goto waste and thus effect how I extract the products).

So I can't think of any really good way to increase the current too much, I am pretty much stuck with what I can get.

 

[willib]

another question.,
what if you added air to the reaction? as in a fish tank airrator (sp) ?
i see the result NaClO3 so it seems to me that you need more O2 in your solution..
also if you strip the plastic insulation off of some wire you can use that to insulate the electrodes from each other thus decreasing the resistance of your cell.

 

[mrjeffy321]

bubbling air through the solution wouldn't help any in this case. The Oxygen is supplied by the water through a long series of reactions. The best way to keep the cell as efficient as possible is to keep the temperature and pH in the most ideal range to facilitate the reaction.

I don't see how insulating the electrodes will descrease resistance. I am using a pair a carbon-graphite electrodes that I am / planning to mount from the bottom of the container sticking up through the solution.

 

[willib]

by insulating the electrodes from each other, ( not compleatly of course)
just using the plastic at either end should do, you would be able to place them closer together..thus decreasing the resistance..

 

[mrjeffy321]

I began using the computer power supply in my newly designed electrolytic cell the other day to see some actual data of how it would preform when used.
I was getting surprisingly high amps out of it, which is a good thing. I was using the +12 V output, and started to get around 7.7 amps. if it had stayed where it began, all would be well, but as predicted, it slowly climed up to about 8.5 amps in roughly 2 hours (not continous, 1 hour on, then 2 hours off, then 1 hour on) of opperation. Everything was getting very hot, the insulation on the wires was beginning to melt, so I stopped it.
Then I tried using the +5 V output on the power supply, this started out very stable, it stayed at about 2.5 amps for roughly 1 hour, then on the 2nd hour, it began to drop quickly, falling down around 2.3 and still falling, so I stopped it again.

As long as the current stays reasonle stable, all is well. I would very much prefer a low[er] stable current to a rapidly changing high current, that is until I get (if ever) my amp hour meter done.
I was thinking, perhaps if I put a capacitor in the circuit, would that stablize things any? Obviously it isn't a fix-all solution, it won't perfectly flatten out a shot circuited cell, or compensate if the current drops too significantly, but it might help.
What do you think?

 

[willib]

if I put a capacitor in the circuit, would that stablize things any?

no , it would be a waste of time..
because you allready have a constant DC voltage , plus you have the current in the supply that appears to be too much at times ..
adding a cap to the circuit will gain nothing..

 

[mrjeffy321]

On a slightly different topic,
I am trying to measure the total charge I have put into my cell, if I know the charge put in (amp hours), then I know how much of the reaction has been completed. Right now, I take reading of the current and then estimate the amp hours for a given time and keep a running total. This is very trouble some, and I'd would like to find an automated way of doing it.
I did have plan on building an amp hour meter, but nothing has come of it, it is too dificult for me to complete myself without someone else helping me.

I got another idea,
What about a watt hour meter insted? This was I will measure the total watt hours through the cell, if I take this number and divide by a constant voltage, it should give me amp hours, right? But I would have to assume a constant voltage, or find the average voltage, for this to work. How would it effect my circuit, I don't know how exactly those meters work, would they dramitcally increase the resistance, or create other undesireble effect to the circuit?
Of course, my first choice is to get the amp hour meter working, but could this work as a good substitute?