The International Open Source Hydrogen On Demand Builders Network.
I've been thinking about building a new cell with parallell plates instead of 5 neutrals more popular design.
A bit about my background:
I've been actrively testing different types of cell for more than 2 years now and the one currently installed on my vehicle is a OGO ( Taiwan ) 25 plates drycell.. Works very well, no corrosion, water stays clear.
The problem is:
The OGO is only drawing 2 or 3 amps, even if I max out the pwm, or add a rediculous amount of KOH.
It makes about half liter per minute or so, not too bad considering current draw..
I use a PWM wich has a fixed frequency of 16.6 kHz ( Kelly motor controller ) and choke coils in serie with the cell and PWM.
Something is wrong here.. Why I'm I only drawing 2-3 amps? I should be able to draw 60A ( and make 10 litter per minutes ) if I max out the pwm ( 12V )
So my idea is to come up with a drycell with a large surface area, possibly doubling or tripling what I have with the 25 plates, and use a smaller number of neutral plates, OR no neutral at all !
This combined with a smaller plate spacing, of 1 mm, instread of 1/8"
I also thought about using a pump, to circulate the electrolyte, wich will be close to distilled water.. less than 1g of KOH per litter.
To resume the design:
HHO flow cell:
-Pump to circulate electrolyte
-Parallel or few neutral connection
-Small gap, 1mm
-Large surface area: 20 plates 4" x 28"
-Tru slots at each ends of the plates, to facilitate electrolyte circulation
-Springs to keep proper compression on gaskets
as for the drive system, I'd use a similar setup to what I'm currently using, PWM, choke coils, except with a variable freq PWM, that has fast switching mosfets, for fast rise/fall voltage.
I think It's better to keep pulse length at a minimum, to let the cell react with the choke coil.. creating a resonnance effect..
For this you need a cell with a very LOW resistance.. thus the parallel connection..
+ you need low electrolyte, to create a more active gas, and avoid scrubbing out KOH vapors..
+ the lower electrolyte makes the cell more capacitive, instead of just resistive, thus increasing the L\C resonnance effect between the cell and the choke coils.. You don't have to be at peak resonnance to see the benefits of it..
Using tru holes ( aligned holes ) in the plates is usually a big no-no in drycell design, due to current leaks created by the edges of the holes, but in this design, the water is flowing rapidly past the edges, due to water being circulated by the pump, avoiding to normaly occuring current leaks.. ( at least that's my theory ;)
Another possible problem with this design would be corrosion, due to higher voltage between plates.
If it's a parallel connection, it'll be 12V ( or more if it is resonanting ) between the plates..
Normally, this would scream heat, innefficiency and corrosion. But with this design, we're using a PWM and a choke coil combination, with small duty cycle, of 20% or less. This, combined with a large surface area of electrodes will helps reduce the corrosion, cause electric stress is spread over a larger surface.. It is also esier to cool the electrolyte with a radiator if the a pump is used.
That said, your comments and ideas are welcomed.
Some comments and questions:
1) How to design and produce those choke coils ? I do not use them...
2) With coils one must use a PWM because the pulses are the key to generate chokes. Right ?
3) It is say that HHO production is more related with pulses then with plain DC. We want HHO at most.
4) It is well established that the gap should be between 1 and 1.5mm for this kind of cell.
5) As it is very difficult to build a PWM that keep the current the same at all temperatures one must use an amp meter to be able to adjust the current via the PWM.
6) There are very few reports telling that pumps help at all the MMW of the system.
Can you explain a bit about those chock coils of yours ?
Hi Thomas and all.
The chokes are simply few turns of heavy gauge ( # 12 or more ) multistrand wire around a ferrite core. I use silicone insulated, multistrand ( 300+ strands, depending on wire size ) just like is often used to hook up RC batteries. They are very flexible, even if the wire is very big.. wich makes winding easier and it stays closer to the magnetic core, wich create the magnetic field more efficiently.
Recently, I've tested both windings on the same toroidal core, but wound perpendicular to each other.. This seemed to give the best results.. This has lowered current draw, and gas output has not changed..
This type of winding configuration lowers the Q factor of the inductor, wich means, it'll resonnate on a wider range of frequencies, but resonnance peak will be smaller. This is good if you don't have a circuit that maintain the cell at resonnant frequncy. (PLL)
Lower current draw may be due to increased impedance with perpendicular windings, or a resonnance at the cell.. In any ways, cell can't reach it's full potential, due to low current draw.. it is still very efficient, but is limited to about 1 litter per minute so far..
That's why I'm thinking about decreasing resistance at the cell, by reducing (or eliminating ) neutral plates.. Hence the parallel plate cell idea..
Those who have tested this idea, myself included, found that it can create lots of heat and corrosion, due to higher voltage between cells. To avoid this, a large surface area of plates is used, and possibly a large volume of water, also, to dissipate heat. Since low electrolyte is used, it makes the cell act like a capacitor, this is what is needed for LC resonnance. Since capacitor creates "voltage lag" when an electric pulse is is applied, it may not have time to rise the the full 12V before the pulse collapse. This is dependant on frequency.. The higher the frequency, the less voltage has time to raise to full 12V.
Since we also have inductor(s) in serie with the cell, it creates a "current lag" wich is a property of inductors...
So, the exact voltage at the cell is an unknown at this point, we'll have to try it out to see if it overheats of not..
You don't have to worry too much aout keeping a constant current on the cell, as long as it does not exceed the safe limit of your alternator. Also, the choke coils act as a current limiter. If resonnance is reached, wich is what we're looking to achieve, it lowers current draw even more, since it's a serie resonnance..
The pumps is used to evacuate produced gas as fast as possible, thus increasing active surface area of the cell.. It may have other effects as well, especially if colloidal metals, such as iron and nickel is present in the water, wich is more than likely with this kind of setup, since the higher voltage has a tendency to detach more metal from the surface of the electrodes. These colloidal metals acts as conductors, moving in the magnetic field from the current present in the cell, the very definition of a dynamo.. So the movement of electrolyte has a "dynamo" effect, wich creates even more gas.. This has been explained in other patents, I can't remember the name, but it dates back from the 80' and had spiral electrodes with circulating electrolyte..
Could you post here a photo of this your choque coil ? I would like to have a feel of dimensions...
I have been doing testes with this kind of coil:
But i do not have the needed electronic skills to really know what am doing.
Wow, nice looking coils..
Mine have lower impedance and a magnetic core..