Ionizationx: a clean environment is a human right!
Projects by members => Projects by members => warj1990 => Topic started by: warj1990 on July 10, 2011, 19:58:47 pm

Wanted to look into / document KOH mix and current draw.
This setup is with a single tube in tube design.
1/2 inch inner tube, 3/4 inch outer tube spacing about 1/8 inch round. 6 inches overlap.
Inner (limiting) surface area: 0.5 * 3.14159 = 1.57 inches * 6 inch length = 9.42 inches
The surface area is equal to 3.07 inch square plates.
I did not measure the KOH put into the cell. I am simply documenting Voltage vs. Current draw.
I have read 28 % KOH by weight is the best mix, for the amount I applied I am well beyond that ratio.
Both tubes were SS 316L material.
The bridge is rated 1000 volts at 100 amps.
Overlooking barrier voltages needed I am putting the effective resistance of the cell also.
VBB = Voltage before Bridge rectifier
VAB = Voltage after Bridge rectifier
A = Current draw
R = Cell resistance (R = V / C)
W = Watts (after bridge source)
VBB VAB A R W
2.79 1.7 1 1.7 1.7
3.1 1.9 2 0.95 3.8
3.5 2.1 4 0.525 8.4
3.7 2.2 5 0.44 11
3.9 2.3 10 0.23 23
5.2 3.1 18 0.172 55.8
Adding from the other day:
3.3 32 0.103 105.6
It is difficult to get good measurements using a Variac at such low settings and minor changes. I did see 40+ amp draw at one point. I quickly lowered the variac before looking at the settings( amp meter and 2 volt meters).
I am thinking that was about 67 volts before bridge.
As far as I know this is the best setup we can do  as all research points to KOH as the best electrolyte for brute force.
Some changes would be to mix the electrodes and gain some voltage the galvanic corrosion chart comes to mind, but I have not tested any other metals lately.
This setup shows the resistance is lowering for each increase, but the power is increasing also.
(question remains why did my driver take so much power at HV if the cell continues to drop resistance?)
Edit added graph of above results.

I'm still working with purified water and have tried an interesting experiment in the hopes of lowering amp draw. My three inch long by 1 inch 304 tubes with 1/2mm gap and with the inner filled with epoxy are set up in a RO water bath. What is new is that I've also added some clear thin plastic tape to the inner negative tube by wrapping it around completely but leaving a 1/8 inch exposed gap around the middle of the inner tube.
Gas production has increased x4 and the bubbles are large and fast with little milky production seen. The tubes and water do get hot but the coil and mosfet stay cool at 100v and 1 amp on the meter. Playing with frequencies and have noticed some better production when I can get the thing to sing and hum on the bench top.
The big bubbles may be boiling water since the current is limited to the area where the tape is absent, although they are evident at 5v and .1a as well as higher power levels.
kb

kb,
I could not find your projects page, if you want I will post a topic here for you to share your progress. (or maybe I just overlooked it on this form?)
Basically what you are doing is increasing the current density to a smaller area. My cell has an open top at present  so I have no way to verify gas production except by amp draw alone, Faraday.
How are you measuring yours to verify 4x production?
Right now I am trying to figure out how to produce more current draw at lower voltages. I am thinking 2 areas here:
1, larger cell size  going super sized on the electrodes should allow more ions between the electrodes.
2, moving to a plate cell and making the gap smaller.
The farther outlook plan is attempt dissimilar metals on the cell.
My overall goal is achieve the low resistance (0.1 ohm or lower) without the need for the higher voltage. I think the SS has a lot to do with this.
Secondly overall goal is try again with the resonating tank circuit, maybe build a better driver that doesn't walk the frequency, but the auto tune is designed for maximum efficiency.

production measured via closed top and water displacement over time, pardon the thread highjack...
I'll light it up this next week and see what we get from the torch. The flame never lies.
I may have been reading your post and left to take care of something and later returned and replied instead of starting a new post..... yah, that's my story

WJ,
Please do not forget one big factor in measuring.
Temperature of water has a great impact on your measurements.
So, when you put NAOH or KOH in water, it will release heat.
1 degree raise will provide in pulling more amps.
And to give you some more advise in this case. I found out that till around 10 till 15 amps, its better to use NAOH.(20 till 25%)
If you go up into higher amps, you better use KOH.(28%)
Both are very nasty chemicals.....my hands sometimes burned like crazy.......No cloves. ::)
Steve

Thanks for the tips.
Putting down one more graph of today's experiment.
This includes the graph from the other day also.
My next setup will be oversized flat SS plates. (verify different surface area VS voltage needed).
My cell setup was not designed for high current, a #4 x 40 screw for outer electrode and #6 x 32 for inner electrode.
As you can imagine I am smoking the connections to these. I need a # 8 wire, good for 50 amps, and I am going with Brass connection bolts.
I did attempt 1 run without KOH it was 4 amps at 88 volts. Gas production did not seem impressive and the water heated up rather fast.
Gas production was improved with the KOH, results of higher amps.
88 volts * 4 amps = 352 watts
(maybe tomorrow I will run a voltage vs amps on just water  for power in right now it doesn't look good).

Wanted to post a thought on this setup.
water is a voltage dependent resistor. The higher the voltage, driving the system, the lower the resistance. So the power input is about a squared function.
As an example
At 2.4 volts it draws 10.6 amps, Power is 25.44 watts. Resistance is (2.4/10.6) 0.2264 ohms.
At 4 volts it draws 37 amps, Power is 148 watts. Resistance is (4 / 37) 0.1081 ohms.
I know what you are thinking, been looking over this all day also. There are 2 things to look at here.
1 the water resistance went down  so the cell can pass more current and be more efficient.
2, connecting 4 tubes together in parallel will achieve the same thing, 2.4 volts @ 42.4 amps. This is still more efficient at 101.76 watts.
Now what IF...
You can achieve a high voltage pulse and deliver several amps low voltage? This would be kind of like a flash circuit where the hv pulse allows the electrons to flow. This is also similar to the plasma discharge spark plugs.
The big difference is how fast the water consumes the high voltage pulse.
The high voltage pulse could be 12 volts @ 1 amp, 12 watts, while the high current is 2 volt @ 20 amps, 40 watts.
Expected water resistance at 12 volts 0.055 ohms (need to verify this with more testing and data)
At this expected resistance the cell is able to handle (2 volts / 0.055 ohms) 36 amps.
So with 2 secondary windings on the transformer and a diode tied in from the high voltage to the low positive voltage we may be able to achieve greater than 100% efficiency. (based on current known methods).
The grounds on both windings would be tied together on one end and sent to the ground of the cell. The positive would be taken to the positive of the cell.
Current is going to take the path of least resistance, which is mostly the water with KOH added.

No experiment time the last few days.
I need to find a fast way to verify the voltage drop  is it due to the movement of water, because of the H2 and O2 causing the circulation of water, or because of the voltage alone.
I have only seen one brief article on water movement and resistance. While it vague on a message board and the question was asked why did the water resistance drop when it was in motion.
So now the question is can I get lower resistance by just allowing the water to flow and not need the high voltage pulses.

Wanted to add another chart.
Large surface area allows lower voltage and more current draw.
Kind of already knew this one, but again documenting some tests.
This again had lots of KOH added.
One disappointment is on the single cell 3 volts @ 20 amps, 4 volts @ 40 amps.
An exact double  the offset results I had when I first started this did not appear.
LATE EDIT  an exact double in voltage would allow exact double in amperage  this only increased by 1 volt and doubled the amperage.
If the effect was not here voltage would have been 6 volts for 40 amps, not 4 volts.
The results are in the 2 and 3 cell tests.
I am pushing the limits of the variac  it is rated at 130 volts @ 20 amps  but keep in mind I am pulling up to 64 amps out of it.
(the overall power is under 250 watts, but really not intended for so many amps drawn at low voltage)

Here is tap water with a single cell.
Gas production rate seemed correct to current flow.
Around 80 volts the graph moves slightly higher as the water is heating up.
My data collection was for every volt, however I did not find a significant resistance change as I have in my other experiments.

I believe Stans circuit with the chokes was to limit the current.
This is looking and related to the KOH and series cell.
At 2 volts the system is drawing between 2 and 5 amps.
At 3 volts the system is drawing between 20 and 45 amps.
The inductor (choke)/ resistor time constant allows the voltage to rise  without the massive amp draw.
It takes time for a choke to allow amp draw to pass through. It also drops the voltage as the current is limited.
So using a choke, verifying the time constant with the cell, and selecting the correct frequency you can apply whatever voltage you wanted and keep the amp draw low.
This prevents the massive current flow with a 1 volt change in the cell.
Here is an example for some clarification.
I have a voltage supply of 0 to 12 volts. I can pulse it on and off 50% duty cycle. The cell will draw 10 amps at 12 volts, but I want the amps to remain lower.
One choke time constant = L / R, inductance / resistance.
Each time constant allows 63.2 % change from total.
We tune the frequency to match 1 time constant of the system. So the choke will have power applied for 1 time constant and off for 1 time constant.
(the time constant (aka frequency) is variable with water resistance and choke size).
Here is the data for the frequency (on/off) and current in the cell (resistor).
Start 0 time.
On 6.32 amps
off 2.32 amps
On 7.18 amps
off 2.64 amps
On 7.29 amps
off 2.68 amps
On 7.3 amps
off 2.69 amps
On 7.31 amps
off 2.69 amps
On 7.31 amps
off 2.69 amps
On 7.31 amps
End.
This shows with the choke time constant and a 50% duty cycle the current will stabilize between 2.69 amps and 7.31 amps.
Remember in some of my experiments the water is a variable resistor  so to keep the current from climbing as we really crank up the voltage we need a pause or break in the cycles. Now we have Stan's pulse wave form.
Keep in mind my system with KOH is going to draw much more amps at 12 volts.
This is all still Faraday based, you apply 12 volts (on and off) but the choke only allows the cell to get 34 volts depending on the time constant, etc...
The only boost you may get is from the bifilar choke  I strongly believe this was just done to amplify the choke value more than 2 separate chokes.
I am about 2 weeks from getting the parts for experimentation on this setup, in the mean time comments are welcome.

Hi WJ,
I also done tests with pulsing and chokes.
You noticed the same as i did.
If you have the right frequency in relation with the bifcoil and the wfc, you will see amps being restricted.
But here also, i could not find more efficiency in the electrolysis proces.
I didnt find any " extra" power kick from the coils.
So, yes. More volts, less amps and no better results.
The only theory i can share was the fact that maybe we can use larger surfaces of electrodes, without shorting the powersupply.
More surface does have better efficiency....
My volts tests stopped at 300volts. Maybe i should have taken it up to 1kv.......
Steve

I think I need to look over your projects more.
At least we have some independent verification, as our results seem to be matching.
I am going to run the natural water tests again up to 130 volts and see if I get the resistance change on a different tube.
Maybe the first tube connected does not have the effect of less resistance with voltage ?
Maybe the water flow rate is what lowered the resistance ?

Here is a chart of each of the cells.
4 different tubes in my current design.
I started with cold water on each test.
Tube 2 shows a little better performance, current draw.
Current draw seems to lower by 1/2 amp with negative in the center and positive on the outside.

I notice the chart is still not a strait line, for current increase. I attributed this to the water heating on the past experiment.
This graph shows the last results with a single cell and I started with cold water at 130 volts, then tested at 50 then at 10.
They are the same current draw  so the lower resistance at higher voltage is not due to heating water  while it is true it is not the effect seen here.

Looking over my past work in this topic I notice the resistance drop around 3 amps on all the tests.
So the 3 amps of gas production is allowing the water to flow, causing a drop in electrical resistance.
So to make our cells more efficient we need to circulate the water (more amp draw for same voltage).
Here is a chart of the last experiment and resistance.
I would consider this to be voltage dependent, however with different water, KOH and Tap, I am seeing different voltage levels(3 volts vs 70 volts) for the effect. While current is showing a pattern, therefor allowing water movement as the gas is produced.
The chart should say "Resistance Ohms  could be associated with water flow/movement"
Not on the current graph.
Wanted to note a late edit on reply # 8 where the voltage raised 1 volt and amperage doubled.
Added Water Resistance VS Voltage, natural tap water.