Site Map
  Site Search



       
Cell Configurator

  Amperage

  Bob Boyce

  Browns Gas

  Bubblers / Tanks

  Buying a Cell

  Calculate MPG

  Cell Assembly

  Cell Spacing

  Cold Fog

  Cold Weather Tips

  Conditioning Plates

  Digital Amp Gauge

  Digital Volt Gauge

  Driving Tips

  Dry Cells

  Dry Cell - Tube

  EFIE Circuit

  Electrolytes

  Engine Animation

  Featured Cells of
  Craftsmen   

  Flashback Arrestor

  Foam

  Gas Saving Tips

  Gasoline vs Alcohol

  Go-no-Go Switches

  HELP & Support

  HHO Pressure Test

  HHO --- Why ?

  How HHO Works

  Hydrogen

  Installation Tips

  Jar Cells

  Magnetic Alignment

  My Dashboard

  Neutral Plates

  NEWS

  O2 Sensor

  OBD I & II

  Plasma Spark

  Plastic Containers

  Pure-Gas.org

  PWM

  Questions

  Safety

  Solar Power

  Stan Meyers

  Tax Incentives

  Water

  Water as Fuel

  Wire Size

  Wire Cell

2nd Amendment

Black Powder Making

Build a Wood Tree Stand

Moonshine Stills

Hemp Revolution

Martin vs Zimmerman

 

Hydrogen Separator Cell

 
"Requires 1 additional gasket per water compartment"

Multiple Cell Stacks require Reversing the position of the gasket hole slots, for each stack. In other words, one stack will have "cut-out" on the left, the other stack - on the right. It is necessary in order to keep the gases separated (collecting in the right channel).
 


Assembled Gaskets & Membrane

 

How it Works

 
Hopefully this information will help advance Dry Cell Technology. A few experimenters are using Nylon Monofilament Mesh to divide water chambers into 2 areas; one for hydrogen and one for oxygen. To get an idea of how this is done, watch the videos below. Notice that they are using 2 gaskets between each set of plates. The Nylon Mesh is inserted between the two gaskets (Red line in the picture above). The mesh keeps the Hydrogen on one side, and the Oxygen on the other side. There are special slots in the gaskets for channeling the gases to separate gas output ports. Instead of HHO coming out of one port, Hydrogen and Oxygen have their own ports. Watch the videos and you will get a better idea of how this works. The method was used by in the early 1900's.
 

Gas Outlet Ports

Notice the gas outlet port on the left. There is a slot cut in the gasket so that the gases on that side of the nylon membrane can escape through the hole. The other side of that hole looks like the hole on the far right. The gasket seals that hole so that gases can not enter on the other side (mix). The membrane separates the gases. The slots in the gasket pass or block the gases; keeping them separated. By alternating the slots, gases can be collected and kept separated. This method can be traced back to, before, the year 1919.

I have tested this method of separation, and it does separate the gases quite well. It also allows the water to refill the chambers. It easily passes through the membrane. It is also important to note that backpressure can cause the gases to mix. That means, we need to make it as easy as possible for the gases to escape through the holes, and also through their respective bubblers. Pressures should be the same in both bubblers. That means the water levels need to be the same. So, it may be necessary to equalize the water levels via a hose between them.

In the picture above, there is a flaw in the gasket shape. It is the small area at the top, to the left and right of the holes. Gases can collect there; you don't want that. If they collect there, it is possible for them to leak through the membrane material and mix on the opposite side of the membrane. We are trying to avoid mixing.

Each of the Outlet Holes is dedicated to only allowing one gas to pass through it. As the gases are made, they rise to the top; the membrane guides them. They are blocked from entering one hole, and allowed to enter the other. This process is repeated as additional electrode plates are added. In the next picture that follows, I am holding a gasket assembly. It consists of the membrane sandwiched between 2 identically cut gaskets. The gaskets are turned to face opposite directions. One faces to the front, the other to the back, so that the "Cut out Slots" are on opposite ends. If you build and assemble your gaskets, individually, like this, then it is almost impossible to make a mistake when you start assembling the layers of plates. Keep the tops on top and the bottoms on the bottom. The sides will always match up.

 
 
 
 
 

Membrane Materials being used:

  • Polyester Monofilament Mesh (works with Sodium Hydroxide, NaOH)
  • Nylon Monofilament Mesh (works with Potassium Hydroxide, KOH)
  • Silkscreen Mesh, T165
  • Uncoated Rip-Stop Nylon
  • 300 cross thread count , per inch, is recommended (or higher).
About mesh size:  Mesh size is measured by how many threads of mesh there are crossing per square inch.  For instance, a 165 mesh screen has 165 threads crossing per square inch.  The higher the mesh count, the finer the threads and holes in the screen.
 
300 cross-counts are recommended, but any count higher than that will work; 400,500; etc.
The mesh forms a thin wall that allows the water to pass through it, but not the bubbles. The H and O ions pass through the water, cross the membrane, and form the gas on the electrode plate they are attracted to (positive or negative). The Hydrogen stays on the negative side of the membrane wall, and the oxygen stays on the positive side. The mesh is a dividing wall; it forms/separates two chambers. The gases rise to the top of their respective side of the chamber, and collect at the top. They escape through the hole in the electrode plate; so, it is very important to make holes in the mesh that match up with the holes in the plates. Cut out holes for the water and for the gas. If you do not, back pressure will be created, and that could cause the gases to mix. The gases take the path of least resistance.
 
Mesh material can be obtained from Fabric shops, Silkscreen shops, Ebay.com, amozon.com, Tent repair shops; etc.

 

1 Millimeter = 1 000 Micrometers
Millimeter is a metric unit and equal to one thousandth of a meter. Spelled as millimetre in most of the countries. Used widely to measure small distances in engineering and machining. The abbreviation is "mm".

Micrometer is a metric unit and usually used for precise measurement of small distances in engineering and machining. The abbreviation is "m".

Gasket Material:

  • High Temperature Neoprene Rubber

  • EPDM Rubber

  • Silicone

End Plate Material:

  • 3/4 inch HDPE

 

 

HHO Separation Cells

   
SC-1 HHO Separation Cell

Global Ecological Solutions - HHO Separation Cells:

  Global Ecological Solutions product web page

             (Exceptional quality and workmanship)
                  "Guaranteed Not To Leak Water"


SC-1  For 12 Volt DC Systems

 * Capable of producing 1 LPM of Hydrogen at 24 Amps.
 * Operate Continuous at 22 Amps / 0.9LPM of Hydrogen

   "Click on the Picture" for more details
 

 

SC-2 HHO Separation Cell  SC-2 is for 12 & 24 VDC Systems.

  * Capable of producing 2 LPM of Hydrogen at 24 Amps/24vdc
    Designed to Operate Continuous at 22 Amps / 1.8 LPM/24vdc
 
 *  Capable of producing 2 LPM of Hydrogen at 48 Amps/12vdc
     Designed to Operate Continuous at 44 Amps / 1.8 LPM/12vdc

    "Click on the Picture" for more details

 

 

 

 

 

SC-3 HHO Separation Cell  SC-3 is for 12 & 24 VDC Systems.

  * Capable of producing 3 LPM of Hydrogen at 36 Amps/24vdc
    Designed to Operate Continuous at 38 Amps / 3.2 LPM

  * Capable of producing 3 LPM of Hydrogen at 72 Amps/12vdc
     Designed to Operate Continuous at 76 Amps/12vdc

    "Click on the Picture" for more details

 

 

HHO-Research - HHO Separation Cells   (Australia)

        (No Longer In Business on the web.)


  Available at http://hho-research.com.au/

 

Video:  Hydrogen Separator Build    
http://www.youtube.com/watch?v=2WUhUFNZY3A

 

 

Hydrogen Fuel Systems - HHO Separation Cell

  *  2 Stacks of 7 Plates in Series

  *  They overstate the LPM for Hydrogen. They say 1.9 LPM at 30 amps.
The cell is not capable of even making that much "HHO" at 30 amps, much      less Hydrogen.

   *  I can not speak for their quality of product or reliability.

  Available at their web site



 

  Hydro Bullet - Separation Cell

  *  I do not have any information on the quality or reliability of their products But all looks well. They have 24 hour customer support.

  *  They overstate the LPM of Hydrogen; so do not expect to get what they say.
Example: They say their 6 plate HHO cell produces 600 MLPM of HHO @ 10 amps,
They say their 6 plate Separation cell produces 700 MLPM of Hydrogen @ 10 amps.

  *  They only use 1 external water fill port, but the water ingeniously gets divided behind the End Plate.

   * They do a good job of Separating the hydrogen and oxygen.

 

  Available at their web site


 

Hydro-gas - HydroGenesis Separation Cells

* I do not have any information on this cell. The web site is very lacking.

* No product information is furnished, what so ever.

* No product review - good or bad.

  Available at their web site

 
 
 
 
 
 

1919 Commercial Filter Press Cell (Separation Cell)

A large number of commercial cells put the anode and cathode comparatively close together, but, in order to obtain reasonably high purity in the gaseous products, a porous partition was placed between the electrodes: this, like increasing the distance between the plates, creates a certain amount of resistance, but it has one advantage of the latter procedure in that it makes for compactness, which is very desirable in any plant and particularly so in the case of electrolytic ones, as one of the greatest objections to their use is the floor space which they occupy. If you study this picture, you see that Dry Cell technology is nothing new.

 
   
 

The membranes go between the neutral and active plates, they aid to keep the gases separate with help of the special gaskets.
They are very accurately laser cut for a number of reasons, they have to fit correctly between the gaskets, and the mating holes need to be accurate too. The edges must not go past the gasket edges or leaks will occur, so the membranes are made 2mm smaller than the gaskets (3mm). It is near impossible to cut these by hand successfully.

   
The cell does not claim to separate 100% but the concept looks promising. I think the hydrogen is going to speed its way to the top...and out. While, the slow oxygen...which is an accumulating bubble with a membrane...slowly follows the contours of the mesh divider.

It may be possible to use this mesh in a bubbler. First, the HHO should pass through the mesh, which would break up the gases into smaller bubbles. Those bubbles would rise in a chamber with another layer of mesh. I think the oxygen gases, which are heavy and slow moving, would linger underneath the mesh, and the hydrogen would pass through it...since it is moving at 20 feet per second. You would need a barb fitting just under the mesh to allow the oxygen gas to exit the bubbler. Plus, a fitting farther above the mesh for the hydrogen. I learned from Joe, of joecells, that the two gases will not pass through the same hole at the same time. With that in mind, it may also work with two output hoses at the top of the bubbler.

There are other theories for separation of the gases. One is to use magnetic fields to help direct the gases out. I know this helps move HHO up and out of tube cells. Stan Meyer used opposite high voltage potential fields. Positive repelled the Hydrogen and attracted the Oxygen.

   
Another Example:   
 Video plays off-site  
           
 
 

 

   
   
   
   
   
   
   
 
 
 
 
 

My concept of the idea:

The membrane is sandwiched between two gaskets.
Hydrogen is collected on one side of the membrane, and oxygen on the other; the two gases are kept separated as they climb the membrane wall..
You will need two top holes for the gases to exit.
Each hole gets obstructed from the gas producing surface of the opposite plate.
Reverse the gasket, on the opposite side of the membrane.
Keep alternating the gaskets this way.

The drawings below depict the general idea. Gasket shape and design are left up to you.

 

  

Below,  is a version using 2 water fills; one for each gas. It would allow better circulation.

 

Separation Gasket & Membrane Assembled.

  • Each consists of 2 gaskets with a membrane sandwiched and glued between them.
  • These were boiled for 5 minutes after the glue dried.
    The heat tightens the membrane by shrinking it slightly.
  • This process makes the gasket easy to handle.
  • It also makes the cell easier to assemble.

 

 

  Ceramic Separation Membrane Development :  The process requires high pressurization of the gases.           

....

    Copyright 2003   All rights reserved.   Revised: 09/18/14.                                             Web Author, daddyo44907
The information presented on this web site is for information purposes only. Should you decide to perform experiments or construct any device, you do so wholly on your own responsibility
-- Neither the company hosting this web site, nor the site designer author are in any way responsible for your actions or any resulting loss or damage of any description, should any occur as a result of what you do.