Gas Cooker & Water Purifier Using Free Energy

I originally wrote this for my final year project in 2005. I won't provide a parts list as certain technology has moved on such as the ease of Arduino for the electronics and solar cells are a lot better and readily available now. Also I was a student at the time and so my budget was extremely limited but even today I am happy with my design.

Many parts of the world suffer from lack of clean water to drink and heat to cook their food and get by from day to day. This instructable is designed to try and provide a cheap means of providing clean water and heat to cook with for a very low price tag.

Hydrogen is very easily made by using electrolysis which can be driven from green technologies such as wind and solar power. Electrolysis will work with stagnant or dirty water to provide Hydrogen and Oxygen gas. This gas mixture is highly combustible and as such makes a great fuel source. The added benefit is that burning hydrogen and oxygen produces 100% pure water.

There are probably lots of flaws with this design and essentially what you are making can be very dangerous if left powered and unattended so please be advised. I've since been employed in the embedded industry for over 8 years and have learnt a massive amount so please if you have any questions (about electronics / microcontrollers / Arduino) then let me know. Anyway I hope you enjoy reading my instructable as much as I enjoyed designing and building it. I do intend to dig this project out the loft and take a video of it in action. Yes it's still working after all this time.

Please read on to find out more....

Step 1: Alternative Energy and the Hydrogen Industry

Well researched methods such as solar, wind or water power are fantastic in that they use the energy present in nature and convert that energy into electricity. So ok with these methods we have electricity being produced so now we need a method to use this electricity in our current technology. This is where electrolysis comes into the picture. Electrolysis was discovered in 1800 by an English man named William Nicholson. The reaction that occurs during electrolysis is simple, two electrodes are placed in parallel and submerged in a liquid, the electrodes can be made of stainless steel, aluminum or any other metal as long as it doesn’t react with the liquid and has a high surface conductivity. When an electrical potential is placed between the electrodes in water the only path the current can flow is via the impurities in the water. At the electrodes electrons pass through the water from the cathode to the anode creating concentrations of oxygen and hydrogen at the anode and cathode respectively.

Electrolysis may seem a simple and useless process to use for energy but when used on water a very unique reaction occurs, where there was a hydrogen oxide molecule there becomes two hydrogen molecules and a oxygen molecule (see Fig 2). As both of these new substances are gasses at room temperature they collect into bubbles and rise to the surface of the water as a mixture. This mixture then makes a readily available fuel source that can be used in an engine, cooker, boat, automobile, airplane, heating e.g. all of the products that run on fossil fuels today such as petrol or coal. One drawback to hydrolysis is when using salt water as this produces chlorine gas which is very harmful. This side effect can be easily avoided by removing any salt content through evaporating the water first. In this project tap water will be used as the means of fuel storage so this problem will not apply.

Since the discovery in 1800 and first documentation in 1832 by Michael Faraday, hydrolysis has been used multiple times to try to produce a new form of energy. However none of the products made have ever been very useful due to problems such as lack of efficiency and speed of gas production. Recently though hydrogen technology and electrolysis has come back into the lime light as the fossil fuel industry starts to collapse. Honda’s newest car for example employs a high pressure tank of hydrogen to power in its fuel cells to power the car.

Unfortunately one of the electrolysis researchers called William Rhodes claimed the invention for himself and named the output gas ‘Browns gas’. He claimed to get more energy out of the hydrogen then was put into the water to create it in the first place. This is obviously against the laws of physics and thermodynamics but still he used his rouse to extort a great deal of money, claiming it was for further research. He then disappeared with the money and was never heard from again however his inventions (scams) still live on.

Another Recent development that has helped the start of the hydrogen industry was the fairly new discovery of hydrogen fuel cells (see Fig 3). These are constructed from two chambers to hold hydrogen and oxygen with a PEM (Proton Exchange Membrane or Polymer Electrolyte Membrane) separating the gasses. When hydrogen and oxygen are present on either side of the material a current is generated, water and heat are given off. However there are difficulties with this technology as when the fuel cell is running one side of the PEM tries to dry out and will crack if not kept wet, the other side needs to be dry to conduct the current and is producing water. This makes it difficult to get the fuel cell to work at all or for any sustained period so it will not be included in the project as the budget and technology is limited.

here is however hope for the hydrogen industry in sight, a company in Australia seem to have got it right, producing goods such as water fueled generators, fuel cells, electrolysis chambers, hydrogen boilers etc, proving that if used correctly hydrogen can be a very useful and versatile fuel medium.

With all this information in mind a project was chosen to demonstrate this remarkable technology and the benefits that it would bring to the fuel and power industries. The aim of the project is to make a cooker to demonstrate how simple it is to manufacture the hydrogen and oxygen fuel mixture locally on site and as needed to prevent the need for storing excess. To demonstrate this technique and test its efficiency the project is going to be based on a hydrogen cooker with gas being produced locally in a sealed pressurized chamber. Figure 4 on the left shows this said system with a device between the input gas chamber and the output hob. This device allows combustion to be quenched if it manages to travel up the fuel pipe before it manages to reach the hydrogen production chamber.

(Fig 5) is a picture of the first electrodes that was constructed to test the feasibility of the output gas as a fuel source. The electrodes were made by sanding down a steel can and shaping a steel metal plate around a bottle so it would fit around the can. Wire was then threaded around the can to insulate it from the metal sheet and act as a spacer to keep the distance between the plates constant. Using a 12 volt battery connected to the anode and cathode of my electrodes I submerged the device in tap water and waited for a result. The output was almost immediate and to my surprise the whole surface area of the steel plates was covered in bubbles of gas given off from the process.

To provide power for my hydrogen cooker system it was decided that green energy should be used because if the system was just plugged into the mains then the process would be running mainly on fossil fuels and would defeat the purpose of this clean recyclable energy. The green energy chosen to power the cooker was primarily solar as this is a well established technology and it is generally always light for some time every day (see Fig 7). A second green energy was then included to complement the solar power and help to keep a trickle of charge running to the battery. This was chosen to be wind power as wind is present all day and night and it is also easy to implement with a simple motor and spinner assembly (see Fig 6).

Step 2: Design Concepts

The cooker system will be constructed as shown in Fig 8. It will consist of a solar cell and a wind turbine powering a rechargeable battery through a regulator. The battery will then power a microchip microcontroller connected to a L.C.D screen and the chamber electrodes. The production chamber will also be monitored by the microcontroller by means of water level, hydrogen pressure and temperature sensors. The sensors outputs will be displayed on the L.C.D and hydrogen production will be shut off after a certain pressure (around 5psi) to ensure safety.

(Fig 9) The first chamber made and designed to produce the hydrogen for the project. After construction it was ascertained that although the chamber was water tight and pressure tight, it probably wouldn’t be very hydrogen tight. Therefore another technique was employed to manufacture the production chamber. Instead of trying to make the whole chamber, bought components with seals would be obtained and used instead to guarantee hydrogen would not escape at least short term.

After shopping around looking for the strongest and best sealed container, it was decided to use a heavy duty straight drain pipe connector and two heavy duty drain pipe end caps. These were chosen because they were strong and resilient and also came with a perfect seal ideal for trapping in hydrogen. These drain pipe pieces are to be connected into the designed system by means of using polycarbonate squares to grip the ends of the two caps (see Fig 10). The polycarbonate will then be squeezed together using strips of threaded metal studding. The controller electronics will then be attached to this studding and encapsulated inside a box to prevent any unwanted sparks getting near to the hydrogen.

With the hydrogen and oxygen mix being produced there will have to be measures put in place to ensure the safety of surrounding population. The worst thing that could happen is if a flame or spark got into the production chamber and came into contact with the pressurized gas. Using a fish tank bubbler submerged in a separate water tank (Fig 11), there is no way for a combustion to travel from the hob to the production chamber.

Rather then a propane or butane cooker which needs a large quench gap, primary and secondary air mixing to burn correctly, hydrogen needs a much smaller quench gap to prevent the combustion reaching the inside of the gas pipe. Also unlike propane or butane, hydrogen cannot have primary or secondary air mixing as oxygen is already inside the mixture. As the gas travels though this quench gap there is a maximum amount that can travel though the hole at any time. Therefore the bigger the quench gap is, the more gas that can come through, making it easier for the combustion to travel back through and along the stream of gas. For hydrogen this small quench gap (Approx 1mm) keeps the system from rapidly loosing pressure and also to help prevent the combustion reaction from traveling back past the gap and further up the pipe. (Fig 12) shows a hydrogen burner with steel wool covering the quench gaps in the piping. This steel wool is useful as it firstly reduces the temperature of the burning hydrogen by limiting the surrounding oxygen from reaching the reaction, and secondly useful by acting like a catalyst. Once the gas is ignited the wool will get heated up and then act to combust any further gas as it flows through helping to create a steady output temperature.

Step 3: The Electrodes

Firstly a second experiment was done by using several aluminum plates and separating them with various insulators to vary the width between the plates (see Fig 13). Each time the width of the insulator between electrodes was altered; the electrodes were clamped together, submerged into tap water and a current was applied. Estimated output was decided on disturbance of water, quantity and size of gas bubbles. After testing lots of insulating sizes and recording the findings, the best result was when the plates were around 1mm apart.

Now that a suitable insulating gap had been found more plates were added to maximize conductivity and spacers were glued to each plate in every corner (see Fig 14). The plates were then held together by means of two pieces of strong pieces of elastic. Each plate was tested for connectivity to its wire and that no connection was present to its surrounding plates.

The wires were then cut to size and secured in place by the elastic (see Fig 15). Bits of the plastic insulation from some mains cable was used to separate the places were the cables touch the electrodes to guarantee that none of them would touch their neighbor.

The finished electrodes were then placed in the bottom of the chamber (Heavy duty drain pipe end piece). Pieces of heavy duty plastic were cut into shape and used to wedge the electrodes in place in the chamber bottom. The whole setup was then glued and allowed time to set (see Fig 16). Conductivity between the electrodes was then rechecked for safety reasons.

Step 4: Bubbler Chamber

Shown here is a fish tank bubbler stone attached to a plumbing end piece (see Fig 17). A hole was drilled in the end cap to allow the hose from the bubbler stone to be pushed through and glued in place. Epoxy resin was used to hold the bubbler stone in place and help seal in the hydrogen.

The finished bubbler chamber with bubbler stone sealed inside is now ready to filter the hydrogen output in order to prevent a combustion reaction from ever reaching the hydrogen production chamber (see Fig 18). Hole drilled in another end cap and a stainless steel pipe fitting was glued in place to allow for the incoming hydrogen gas to get out again and be piped to the hob.

Step 5: Pressure Chamber

With the production chamber partially completed it was time to test the output from the electrodes. The chamber was partially sealed to contain the produced hydrogen and oxygen gas and then a filled with tap water for the chemical reaction to occur. A 12 volt potential from a small lead acid battery was then applied across the anode and cathode of the electrodes. Output gas from the chamber was piped into a container of water and allowed to bubble into an upside down submerged container (see Fig 19). A stop watch was then started to measure how long it took the electrodes to produce 250ml of the hydrogen and oxygen mix. After only 11 minutes of producing gas the 250ml level was reached and the electrodes were then deactivated. The hydrogen and oxygen was then tested for its purity by raising the container up out of the water at the same time as inserting a lit splint into the gas. (Warning... I don't advise you do this unless you know what your doing!!!). A rather loud ‘pop’ came from the mixture as it combusted proving that there was indeed hydrogen and oxygen present in a substantial enough level to provide power for a device. The bubbler chamber was not needed for this test as currently no hydrogen or oxygen is being held and allowed to build up anywhere to any hazardous (explosive!!!) pressure.

(Fig 20) is the completed hydrogen production chamber with analogue safety pressure gauge attached. This extra gauge is needed to a: verify the digital input pressure and b: act as a backup if the microcontroller fails or hangs. Also shown on the right of the chamber is the release valve used to control the flow of the output hydrogen and oxygen.

Step 6: Electricity Production

In order to construct the solar panel, flexible bare-bone solar cells were obtained from cps-solar.co.uk, and then glued to a square sheet of Perspex with metal studding and nuts added at the corners to make adjustable legs (see Fig 21). The adjustable legs allow the panel to be faced towards to sunlight from just about any location. Each panel was then connected in series to provide to maximum voltage possible and a standard low noise and low attenuation phono cable was attached. When tested in sunlight with open circuit connection voltage across the cells was upwards of 20V. Testing with the voltage connected across a load of 1K Ohm allows for the power of the solar cell to be calculated.

Power in Shade = 20mW

Power in Light = 53mW

Power in Sunlight = 145mW

The wind generator was made by simply mounting a motor on a small square of Perspex with the shaft facing through the Perspex. A wind spinner was obtained from Ebay.co.uk and attached to the motor shaft by means of a bored out gear. A second larger square of Perspex was then used for the bottom of the spinner to rest on, allowing for the motor to spin with less friction as the weight of the spinner is no longer resting on the motor brushings. Legs were then cut from steel piping and glued into place to support the spinner in a strong wind (see Fig 22). It was then found that the spinner could spin both ways in the wind which means that the motor would either be charging
or discharging the circuit. This is obviously not acceptable so a diode rectifier was used on the motors output voltage to provide a correct positive voltage to charge the battery. A second problem was that the wind blowing on the spinner would cause it to accelerate and decelerate as the direction of the wind changed and therefore the direction of the spinner. This was solved by placing strips of laminated card down the legs in order to channel the wind and make the spinner rotate in the preferred direction (see Fig 23). A low noise and low attenuation phono cable was again used to output the voltage.

Now to implement the constructed hydrogen and oxygen production chamber two circuits have to be made. The first circuit is going to take current from the solar cells and wind turbine and then allow a lead acid 12V battery to be charged. This circuit will also provide power to the second circuit. The second circuit will contain a microcontroller to handle all of the sensor inputs (Pressure, Temperature, Water Level & Battery Charge), a L.C.D to display the sensor data and a relay to control the charge on the electrodes.

Step 7: Circuit 1 Battery Management

Before the first circuit can be made a decision has to be made as to whether to use a series regulator or a shunt regulator to control the charging of the battery. These devices are necessary because they contain a sensing circuit allowing for the current flow into the battery to be regulated. In a series regulator (see Fig 24) current is switched off when the voltage rises above a preset level. In a shunt regulator (see Fig 25) rather then switching off the power, the excess power is spent through a separate load.

Although series regulators are simple to implement they have two disadvantages to shunt regulators. These disadvantages being that series regulators do not use all of the power produced to go on charging the battery and allow the battery level to drop to under charge status before switching the charger back on. Due to these disadvantages a shunt regulator will be used to control current flow into the battery from the green energy sources.

The schematic for the shunt regulator circuit (Fig 26) allowing the solar and wind generated current to charge the battery. When the battery is charged up to 13V the current is disconnected from the battery by means of a 1 Ohm power resistor load and a red L.E.D lights. Then when the battery voltage drops below 10V again the shunt regulator once again connects the input current back into the battery again. The circuitry employs the use of a 555 to allow for sampling of the battery level rather then a continuous feed which would waste more current. A green L.E.D lights when the circuitry is feeding current into the battery. The circuit also allows for the input battery terminals to be connected via wires into the other microcontroller circuit previously described.

The printed circuit board was designed by looking at the size of each component to go into the circuit and making pin templates to match their connection on the board. Then when all of the components had a pin layout, the components were arranged and rearranged around each other to allow for the least connection to have to be made. Whilst doing the rearranging and re-wiring of the circuit, attention was paid to the reverse orientation of any Integrated Circuits present in the circuit. This is due to the fact that the circuit is on the other side of the P.C.B to where the components are. Once a suitable circuit diagram had been made the diagram was cut out and lightly stuck onto a piece of copper clad fiberglass board. The copper clad board was then cut down to size and 1mm holes were drilled through to allow component pins to fit through and be soldered later. Next the circuit diagram was peeled back off the copper surface and after a bit of a sanding to remove any impurities from the copper surface the circuit tracks were drawn on with a etch resistant pen. The board was then left for twenty minutes to dry before being submerged into a bath of hot diluted ferric chloride hexahydrate to dissolve away any unshielded copper on the board. After another twenty minutes the board was removed and rinsed well with cold water to purge any remaining chemicals. Another quick rub with some sandpaper and the circuit board is complete and ready to use. After soldering all the components into place on the P.C.B Wires were attached to allow for flying L.E.Ds and input power connections. A small plastic case was obtained and drilled to allow the L.E.Ds and phono attachments to fit into place (see Fig 27 and Fig 28).

(Fig 29) is the completed shunt regulator attached to the chamber as described above (Fig 10). The red wire that can be seen inside the larger box is one of the two output wires from the shunt regulator to power the control circuitry and the electrodes. Also shown here is the bubbler chamber fastened to the side of the hydrogen production chamber.

Step 8: Circuit 2 Diagnostics & Control

To begin on the control circuit a few decisions must be made to decide how the data is going to be processed. These alternative configurations are as follows: A. Use a simple Pic microcontroller controller to address multiple remote A.D.Cs and use an A.D.C for every sensor input. B. Use a simple Pic microcontroller with one remote A.D.C and use remote relays to switch between the sensor attached to the A.D.C input. C. Use a slightly more complex Pic with a built in A.D.C module to input all of the sensor readings straight to the microcontroller.

For the project it was decided that option C of a Pic with built in A.D.C was a better choice due to less circuitry involved and therefore hopefully less things that can possibly go wrong in the manufacturing stage. The Pic that was chosen was the 28 pin 16F870 as this provided up to 5 A.D.C channels with 21 input/output pins in total. This was also chosen because when running at a clock speed of 4 Mhz the actual program speed is 1 million instructions a second. This is obviously more than enough to control a simple L.C.D screen, a few sensors and some electrodes.

The first circuit to be designed for the Pic 16F870 was just intended to allow the microcontroller to run and be available for in circuit serial programming or ICSP (see Fig 30). The circuit was designed by following the instructions for timing and electrical setup on the chips datasheet.

The next stage of development for the circuit was to build a LCD interface to the Pic16F870. The LCD that is being used is a 20 character by 2 line display with optional backlight (see Fig 31).

The LCD is interfaced by a 16 pin connection where 11 or 7 of the pins need to be interfaced directly by the microcontroller. Out of these 7 or 11, 1 is chip enable, 1 is read or not write, 1 is data pulse enable and the other 4 or 8 pins are used for the input/output data bus. In order to avoid using too many of the microcontroller input/output pins these 7 pins can be condensed down to two pins by using a simple I.C. with 4 flip flops, 74LS174 (see Fig 32). The other 5 pins on the LCD are for ground, input voltage 5V, Contrast control via the 10K potentiometer and the last two pins are used for connection to an inverter to power the backlight. With the LCD working it was ascertained that the backlight was not necessary for the LCD to operate correctly so an inverter was not obtained to power the circuitry. The 74LS174 in the circuit below operates by taking a serial input of data and then shifting this data throughout the flip flops until there is four bits of information to pass to the LCD via its four or eight pin data bus. When the last of the data is loaded in the reset pin becomes high canceling out the effect of the 1n914 diode causing a higher voltage on the data bus then usual. When this abnormal voltage level occurs there is current left to carry on to the enable pin on the LCD to clock the four bits of data into the LCD’s memory.

By combining the 2 pin LCD circuit with the ICSP circuit and connecting the data and clock pins to pins C4 and C5 respectively we now have a P.C. programmable LCD controller (see Fig 33). The Pic16F870 was programmed with the assembler code for 2 pin LCD (see Software Listings) via a Velleman K8048 PIC programmer with ICSP output header. With the LCD circuit working and displaying text the controller circuit is now ready to take in sensor readings and output the data via the display. To send the data to the LCD display the microcontroller needs to send characters in form of two hex numbers or 8 binary numbers.

To implement the sensors the microcontroller must use its on board ADC module to interpret the values passed into the ADC enabled ports. These ADC enabled ports are the 5 pins of the A port on the Pic16F870 microcontroller. The ADC module onboard the microcontroller is controlled via the use of several eight bit registers embedded inside a few of the chips memory locations. These special ADC registers are referred to as ADcon0, ADcon1, AdresL and AdresH. The ADcon0 memory register location controls the operation of the ADC, the speed of the comparison and the triggering of the conversion process. The ADcon1 register controls the addressing for the output pins and control to switch on or off the reference voltage. For the project ports A0 to A2 were enabled with the ADC function to provide input ports for the temperature, pressure and battery level. Also in this ADcon1 register is the option to justify the 10 bit answer to the left or the right hand side of a 16 bit register. The 16 bit register is really made from two 8 bit registers inside the microchip which are the previously named AdresL and AdresH. The reference voltage was set at +5V or VCC this is always going to be stable as long as there is power to drive the microcontroller as it is being supplied from a 12 volt source through a 74LS05 voltage regulator.

The Temperature sensor or thermistor was configured by consulting the device’s datasheet and looking at the temperature to resistance graph. Three values of temperature were chosen to configure so that their output onto the ADC would give a readable measurable value similar to the data sheet. This was calculated by dividing 5V by the total resistance from 5V DC to ground. The required voltage is the divided by the previous number calculated. As can be seen in Fig 34, resistances for the thermistor have been recorded at several key values in order to calibrate the input circuitry. The thermistor was then connected to the controller circuitry by means of a twisted pair connection. This type of connection was chosen because of its high immunity to noise and its low cost to put into practice. To create the twisted pair, the two ends of a long wire were taped to the rotating shaft of a hand drill. The middle of the piece of wire was then secured, allowing the wire to be wound around itself. By keeping the wire slightly taught at all times the twist of the wire can be made uniform which helps to increase the cables noise immunity. A value for half way through the exponential response of the thermistor was chosen at around 145 Ohms. The error can then be calculated by the microcontroller to give an accurate reading.

R uC = V uC / (VCC / R total)

The Battery sensor was constructed in a similar manor to the temperature sensor with the total voltage varying instead of the resistance. The 12 volts passed from the shunt regulator is variable depending on the charging status and the charge in the battery. The 12 volts was then connected across a potential divider circuit to bring the voltage down to a level readable by the ADC on the microcontroller. To be able to give accurate readings of the voltage on the LCD the resistances again have to be calculated and recalculated to provide a low voltage that is a linear representation of the input 12 volts. Seeing as the system voltage can only really go between 14 volts at a max and 8 volts as a minimum these are the values that were used to calculate the required potential divider resistor values.

The pressure sensors however were not so easy to configure and proved to be quite troublesome towards getting the project working. The circuit diagram for connection taken from the device datasheet (Fig 35) showing the four pin pressure transducer in a wheatstone resistor bridge circuit. From the diagram it was taken than the device came hosting the bridging circuit and it would be a simple case of connecting the positive and negative Vo to the input positive and negative terminals on an op amp. The device powered up and seemed to be working as I gave an output to the ADC. However the output value to the ADC from the pressure transducer was fixed and didn’t vary at all with pressure change. On the second look at the datasheet circuit diagram (Fig 35) it was decided that it was possible that the device maybe did not have built in bridge resistors. A circuit was built up around the pressure transducer to perform the balanced bridging function of the wheatstone bridge. The output of the bridge circuit was again fed into the inputs of an op amp to give the difference of the positive and the negative as a voltage. When the pressure transducer was powered up in the bridge circuit it again fed a fixed constant number to the ADC indicating that either the part was faulty or there was something further to the electronics that was not fully comprehended. Every other attempt to interface the pressure transducer caused the microcontroller to crash or run in a very bizarre manner.

Due of the above difficulties with the digital pressure sensor it was not possible to include it in the project. However with the backup analogue pressure gauge the pressure inside the production chamber can still be monitored accurately so it is not a great loss to the systems overall performance.

The next device to be connected to the controller circuit was the level sensor. This consisted of a pole with a magnetic float that could move along the length of the pole. On either end of the pole were caps to stop the float from coming off, and reed switches to sense the magnets proximity. Both switches are open circuit when the magnet is in the middle of the pole. Using a simple resistor above the value of 4.7K Ohm connected from the microcontroller data pin to ground allows for extra current from the switch to flow around the microcontroller without causing any damage. This resistor is also important as a device to load the input voltage across causing the voltage to be determinable at that junction. The controller software was now upgraded to include all of the sensing devices (apart from the digital pressure sensor) so that the ADC values of each sensor were read alternatively and the data from the sensors was output to the LCD.

Table 1 describes the process of each input / output pin used from the Pic16f870 in this project.

Step 9: Hydrogen Hob

The hydrogen burning hob was constructed using the research of several alternative energy websites, magazines and articles. Shown below are two pictures of the finished hob (Fig 38 & 39). The hob was constructed by obtaining some stainless steel piping and cutting it to form a C or a U shape. The ends were connected together with pneumatic couplings designed to push on and then hold a great deal of pressure before coming back off. The two ends at the top of the U were sealed using screws coated in epoxy resin. Five 1mm gas outlet holes were then drilled into the longer pieces of tubing along their approximate center and all facing the same way. Steel wool was then wrapped around the piping to cover the outlet holes and allow for the catalyst behavior as previously mentioned. The whole piping setup was then attached to a sheet of aluminum with several large heat sinks attached to help dissipate the output heat. An output pipe made of stainless steel was then attached to the t junction at the bottom of the U which then plugged into the pressure proof tubing obtained to contain the hydrogen.

(Fig 40) is the sealed hydrogen and oxygen production chamber with electronics and hob attached and ready to be tested. The yellow block that can be seen inside the controller circuitry casing is the connection to the ICSP header on the Pic programmer.

Step 10: Finished Unit and Conclusions

Shown here on the right (Fig 41) is the completed hydrogen generator with hob attachment and LCD interface. Labels also added to the front of the shunt regulator to make it easier to tell the plugs apart.

Now that the project has been completed it is ready for testing and experimentation. From building the project I now have a much clearer understanding of pressurized systems and of hydrogen production itself so in a sense to me the project was a complete success. I also now have much better knowledge on the workings of LCD and ADC controllers, calibrating sensors and making PCBs to a standard. I have enjoyed working on the project and to me it has evolved lots of ups and downs with things going right or in some cases very wrong. I have completed all my objectives and milestones set in the specification. Since the project would appear to have been a complete success with nothing too major going astray. However there would be a few things that I would change to improve the project were I to restart over. I also believe my project was worth the time and effort because it helped me to get a clearer picture of what to do and not do when working with a pressurized system dealing with explosive gas.

Starting again now with all of the useful things I now know about hydrogen technology and making reliable systems would allow me to include all of this knowledge to be able to put it all into the design stage. Rather then how I worked this year doing research and finding out lots of new technologies and solutions to problem while some of the work had already been done. I would spend more time researching everything before hand so that I could start the project with a clearer idea of how to proceed. However I am happy with the construction of the hydrogen production chamber as it is strong and durable and I am also happy with the construction of the green energy systems as they both work well and provide an adequate amount of power.

Before the project demonstration I intend to have answers to the following questions by testing my chamber with a few trial runs. The questions are:

How long do the electrodes need to be on for enough gas to burn for 15 minutes?

How long does battery last with electrodes on full power?

How long does battery last with charging factors helping to maintain charge?

How much gas does the water inside the chamber produce?

Once the system is working correctly there are still many thing that could be used and tested to optimize the system. For a start the extra power that is created by the green energy could be connected across the electrodes when the battery becomes full. Also if the signal going to the electrodes was A.C. then the electrode plates would charge up as anode and cathodes alternatively. This A.C. frequency could be tuned to the electrode plates to create resonance similar to an inductor capacitor resonator circuit.

If there was a proper lab to do research and put work into then spiral electrodes could be made similar to my rather crude attempt in Fig 42. This was my attempt at a spiral electrode and though it looks good it wasn’t anywhere near as tightly wound as I imagined it could be. If two matching spirals were made that could slide in and out of each other leaving a 1mm gap on all sides then this could be used to vary the amount of electrolysis occurring and act as a throttle for the gas production. Also the fact that the material is a spiral means that as much of the surface area is taken up with the electrode material as possible creating the best possible contact with the water.

Another recent addition to hydrogen and electrolysis research is a new cultivated bacteria that was designed to eat any waste by product out of the water that electrolysis will no longer work on. As the bacteria eat the waste product they break down the remaining water and produce yet further hydrogen and oxygen. This helps to supplement the voltage induced electrolysis to produce more gas faster and with less energy involved.