I have a very simple objective, yet a reasonable solution eludes me. I need to convert a 12V DC source to 2V (or close), however I will be drawing around 25 to 30 amps from that source. I need a very efficient yet relatively cost-effective way to divide this potential.
EDIT: Another thought... Is there a simple way that I can use the 12v source as current source, and mix it with a 2v potential?
Wow... what are you trying to do?
Take a peek at http://www.vicr.com/products/dc-...
AFAIK they do all sizes powers etc of power supplies.
It is intended for a current source in a water electrolysis tank for atomic hydrogen and oxygen production. I COULD use the 12V straight, but it would produce excessive heat in the tank, figured it would be easier to dissipate it outside of the tank.
I have looked at some DC converters, but at those power levels they seem to be VERY expensive. Maybe I could bridge together multiple regulators? I will take a look at the link you provided...
Hmm, instead of one monstrous tank with 2V @ 30Amps, couldn't you connect more smaller ones together? Or more big ones together.
Well fact will remain, I still will need about 10 amps per sq. in. of electrode plates, so making multiple tanks would actually be counter productive... ;)
Well, it looks like it's time for a buck SMPS, with high efficiency in the recovery diode (substituting it by a MOSFET). Check out Maxim's SPMS controllers and design tips. Perhaps a 3V3 @ 10A will fit the bill for you. And you probably can even have few of them paraleled.
Anyway, there are few guys in this forums that are quite skilled with SMPS's, and probably they will give better information than I.
This is just one option
http://www.linear.com/pc/downloa...
I think every major SMPS-controller manufacturer
offers solutions in that range.
Very good indeed ossi, TY for that link... It seems by far the best choice right now. Previous contender was
http://www.micrel.com/_PDF/mic21...
But the Linear is the same price and rated at twice the output power. I figure two or three of those should handle my load and leave some space for expansion...
Again thank you all for your help.
Watch out for the minimum output voltage on buck switchers. That is limited by the internal reference voltage. A chip with 2.5V at the feedback node cannot go below 2.5V without playing some nasty games.
Because of the increasing demand for low-voltage processor cores and such, there are more switchers with lower output voltages. Just watch out for what you choose.
Jim
Agreed Jim. However the output voltage is not critical for my application, it's just that 12V is excessive. I could work with 3.3Vor even 5V.
take a look at an old PC motherboard, might give you everything you need for a quick test.
What is driving the 12 V source, and do you have to go through that stage?
Any link to the theory of electrolysis, with data and graphs for the impedance of the system as a function of the driving voltage, (assuming the same electrolytic solution and electrode surface area)?
A big tank of water seems like the ideal heat sink for any heat generated by the system. Heat generated / dissipated outside of the tank still has to be dealt with, (heat sinks, fans, etc.) That energy also contributes to the system's inefficiency, it certainly doesn't go towards gas production.
Just a thought or two, albeit from someone who hasn't done any water electrolysis since school, and that was on a mighty small scale compared to what you are describing!
JC
Um, if you only want a couple of volts across your electrolysis tank, why not put multiple tanks in series? Same current flows through each, voltage is divided across each, no power is wasted. Or Mr Kirchoff was lying!
If excessive heat dissipation is the only concern, why not use PWM?
What is driving the 12 V source, and do you have to go through that stage?
The source is an alternator. Very noisy, minimal regulation.
Any link to the theory of electrolysis, with data and graphs for the impedance of the system as a function of the driving voltage, (assuming the same electrolytic solution and electrode surface area)?
Unfortunately, because this method is not widely used, even though it was discovered almost a full century ago, there is not much scientific data to go by. As of today, less than 4% of the world production of hydrogen is made from electrolysis. The vast majority of the commercially and industrially-created hydrogen and oxygen is made from hydrocarbons, which results in much more impurities in the final product. However digging around I have collected SOME information.
Using perforated 316 stainless steel electrode plates (like this: http://www.mechanicalmetals.ca/p...) with 40% of the surface being open area, roughly 10A per sq. in. of polarized plating is required, plates being separated by 1/2", and using a 7% sulphuric acid solution as dielectric. For the ionization reaction to occur at full potential, only a voltage of about 2V is required. Any further energy will be dissipated as heat.
A big tank of water seems like the ideal heat sink for any heat generated by the system. Heat generated / dissipated outside of the tank still has to be dealt with, (heat sinks, fans, etc.) That energy also contributes to the system's inefficiency, it certainly doesn't go towards gas production.
Normally I would agree with you, however keep in mind that this is going to be under the hood. Heat will already be present to add energy to the reaction, I certainly do not want to add any more unnecessary heat to the tank. My other concern is that I need to monitor tank level, temperature and current, and adjust current limit to the electrodes in consequence. There is no easy way that I can do all this without a proper power supply.
Considering full power being obtained @ 5V 50A (which is grossly overestimated; I do not think I will go over 30A), that is 250W total power, about 25W will be dissipated by the switcher, most of it by the output FETs, which I can easily take care of with a couple heatsinks attached to a stainless steel enclosure, no fan other than the engine's.
Just a thought or two, albeit from someone who hasn't done any water electrolysis since school, and that was on a mighty small scale compared to what you are describing!JC
I appreciate them! ;)
If the power source is an alternator, you can use a transformer to step the voltage down before rectification.
Um, if you only want a couple of volts across your electrolysis tank, why not put multiple tanks in series? Same current flows through each, voltage is divided across each, no power is wasted. Or Mr Kirchoff was lying!
Simply because then I have to build many of them, and will be missing some space. Also because I need some other capabilities from the power supply I cannot get just by putting them in series, like current-limiting.
If the power source is an alternator, you can use a transformer to step the voltage down before rectification.
Hmm not a bad idea, but that means i would have to open up the alternator, connect a tap before the diode pack, and buy a transformer that is likely to cost me over 100$ for 50A rating. At that price I might as awell have the alternator's rotor rewinded to output the voltage I need for less than a hundred bucks... ;)
Are you going to burn the gases in the same engine that is providing the power to electrolysise the water?
Is electrolysise a real word?
I would use a simple buck regulator with synchronous rectification for cool operation.
OK, are you building a perpetual motion machine? If so, you will need a very, very efficient buck regulator.
Are you going to burn the gases in the same engine that is providing the power to electrolysise the water?Is electrolysise a real word?
Cool huh? ;) The gases will supplement a regular gasoline engine, via the air intake. Trick then is to adjust fuel-air mixture a little leaner, and let the O2 and H2 do the rest. This has the benefit of burning perfectly clean, also it burns at a higher temperature then regular gasoline-air mixture, and cleans carbon deposits in the cylinders and valves. Upon combustion H2 and O2 recombine to form water vapor. If this was a hydrogen/oxygen engine ONLY, you could take the exhaust, cool it down, and recycle it to the fuel cell. If this first one ends up working as expected, next step will be to retrofit a propane engine.
I would use a simple buck regulator with synchronous rectification for cool operation.
Indeed that is the plan. I was hot for the LT1339 for most of the day, but finding the external components to accommodate a single 50A source proved to be quite a hassle. Now I have found this one:
http://datasheets.maxim-ic.com/e...
Over 90% efficiency @ 3.3v 20A, 12V in, and made for parallel operation! :)
OK, are you building a perpetual motion machine? If so, you will need a very, very efficient buck regulator.
Not a perpetual motion machine no, we all know that doesn't exist on earth... ;) But a more economic diesel or gasoline engine, sure!
There is a lot of very dangerous devices sold on the internet to do just that, most have no adequate protection in case of backfire, and are made junkyard style. They caused a lot of the bad publicity this kind of system has gotten in the past, people blowing stuff up because of stupidity. Stuff as simple as using 18g wire to drive a 30A load. Or using a solid flashback stopper instead of a water bubbler (Brown's gas has too high a flash velocity for any solid stoppers to be effective; faulty hose ignites, flash goes back to the electrolysing tank, where anywhere from 1 to 5 pounds of gas might be waiting...). Or driving 12V @ 40A and letting the electrolyte levels go low enough that two plates too close to each other arc.
Efficiency is indeed the goal here, be it just for the sheer amount of heat an inefficient power converter would produce.
It seems to me that the amount of power required from the engine to generate the H2 and O2 would be significantly higher than the amount of power derived from burning the H2 and O2 in the engine.
It seems to me that the amount of power required from the engine to generate the H2 and O2 would be significantly higher than the amount of power derived from burning the H2 and O2 in the engine.
This is a big point of debate on the subject, however I have no definite answer for it. Using a switcher as power supply, converting 12v to 3.3v @ 90% efficiency for a load of 30A (108W at converter output), results in a +-8A load on the alternator/battery, + quiescent current of the converter. Can I produce 15 watts of supplementary power from the gases and thus reduce the gasoline consumption? I think so. I guess test will provide proof... ;) EDIT: I think this should be more regarded as a transfer of power (from the costly gasoline to the cheap water) rather than in terms of efficiency. The loss of power through the creation of the gases gets compensated by the lower price of fuel for the same mileage.
As an example, how much more gasoline does your car use up when you drive around with the stereo volume loud rather than quiet? Or does your MPG go down when you drive at night with the lights on than during the day?
I use Power One POL's for multiple designs. (10 amp version) but it seems they have a 60 amp @2.75V version. The cost of one module might be a little $$$.
http://www.power-one.com/board-mount/digital-power-management/no-bus-digital-pols.html
This is a big point of debate on the subject, however I have no definite answer for it. Using a switcher as power supply, converting 12v to 3.3v @ 90% efficiency for a load of 30A (108W at converter output), results in a +-8A load on the alternator/battery, + quiescent current of the converter. Can I produce 15 watts of supplementary power from the gases and thus reduce the gasoline consumption? I think so.
Are you thinking that you can produce over 125W of additional power in the engine, (108W for the converter, 15W of additional power for propulsion, plus covering any losses in the various conversion processes) simply by adding some additional hydrogen and oxygen into the engines air/gas mixture? Because that's what it's going to take... in order to get 108W out of the converter to produce hydrogen, you are going to have to put at least 108W into it. I fear you're chasing a pipe dream... while you might increase the engines output slightly, the increased load on the engine will far outweigh that increase, resulting in lower performance overall, not higher. If it was the other way, you truly would be getting more work out than putting in, thus creating a perpetual motion machine.
Not meaning to discourage you... just discussing the topic. By all means go ahead and experiment, because often that is the best way to learn! :)
Just in the general interest of producing explosive mixtures, when electrically cracking water to produce the H2/O2 mixture, why even use DC? Wouldn't AC work just as well by alternately forming H2 and O2 bubbles on the electrodes?
Although I agree with glitch and Tom's previous posts, at least when the enginen is idle at some traffic light, or when braking, then the electric energy can be used to generate H2 to use it when the traffic ligth is green again, or when one has to accelerate. Adding alternators to the wheels that will produce electricity only when pressing the brake pedal will act like KERS. So this will definitively recycle some wasted energy.
BTW, all the theory and data (only 4% of H2 obtained from water) posted here should be feeded to all those blind ecologysts that defend the H2 engines.
I'm very intrigued also by the efficiency of the electrolisys by itself. I mean, how much KWh it needs to generate one pound (or Kg or m3 or whatever) of H2, and how many KWh can we obtain from this amount of H2. I bet that even with high performance fuell cells, simple NiMH batteries would do the trick with much higher efficiency, thus rendering any H2 defenders to be nothing more than big oil companies dogs.
A simple gas or diesel engine never will have higher efficiency than 30-33%, the rest being dissipate at the heat excanger. Mos usual engines are in the 25% range in fact. With H2, I seriously doubt that this efficiency will increase too much.
Just in the general interest of producing explosive mixtures, when electrically cracking water to produce the H2/O2 mixture, why even use DC? Wouldn't AC work just as well by alternately forming H2 and O2 bubbles on the electrodes?
Using AC, electrons would constantly alternate from one direction to the other, I am not sure this would leave enough time for the molecules to split and polarize properly. Since the goal here is to remove electrons from the water molecule, I have a hard time imagining how changing polarity 60 times per second would force electrons to one or the other of the electrodes. I imagine it might still be doable, albeit with a very low frequency AC signal, and will probably require mroe energy for the same volume. Kinda like digging a hole but stacking the dirt on the walls instead of throwing it outside the hole.
UNiXWHoRe, can I interest you in my system of turning base metals in gold, winning at roulette and harnessing the magical healing powers of the pyramid? I accept Paypal and Western Union credit transfers. You'll have the last laugh, as you cruise by in your pyramid-shaped sold-gold car, waving your big wad of winnings!
Here's some more data I just came by. Chemical energy potential for gasoline is 45.8 MJ/kg, for hydrogen 142MJ/kg. So essentially replacing part of the gasoline in an engine's fuel mix with hydrogen should yield about 3x the raw energy for replaced amount.
Let's say we have a gasoline engine that burns 1 pound of gasoline per hour at a set speed with a air:fuel ratio of 14.7:1. That means that for every pound of gasoline burnt, 14.7 pounds of air is burnt. Assume air composition of 79.5% nitrogen and other inerts, 20% oxygen, and .5% others, including about 0.5PPM for hydrogen. 14.7 pounds of air thus contains roughly 0.0000294 pound of hydrogen. Brown's gas contains H2/O2 in a 2:1 ratio. This means that one pound of gas contains roughly 0.6 pound of hydrogen. Assume one pound of gas can be produced and consumed in an hour. This changes the air intake composition to somewhere around 4% hydrogen, 22% oxygen. If I can reduce gasoline in fuel-air mixture by about 4%, I will effectively release about 8% more energy from the hydrogen, minus mechanical losses.
I am by no means a physicist or expert in thermodynamic, and I am almost certain there are problems with my reasoning, however the gross estimations seem about right to me, and seem conservative compared to some of the estimates out there on the net (some claim results as high as 40%-50% more MPG on older model Honda Civics; exagerated? most probably. Totally fake? I doubt it.) Now does pulling 8A @ 12V from the alternator result in more than a 4% increase in gasoline consumption?
Well, I will bet that to extract one Kb of Hydrogen, you will need far more than 142MJ of electrical energy. But to obtain this energy, the engine should spend a certain amount of gasoline, that probably wouldn't be burned if this extra energy demand wouldn't exist.
Thus the only way to not increase fuel consumption to obtain this H2, is using electricity generated by the car when braking, instead of waste this kinetical energy trowing it to the air as heat. KERS systems came to my mind again.
Thanks for the info, by the way.
Guillem I agree with you that a boost can, and will, be seen if one uses braking energy (or any other form of "lost" energy). I doubt any would come from generating while at idle, as you are placing additional load on the engine, thus it will need to burn more fuel to maintain idle. (the system is always in equilibrium)
Too bad thermo-electric generation isn't more compact or efficient, as that would be another great source of energy from a combustion engine.
I am by no means a physicist or expert in thermodynamic, and I am almost certain there are problems with my reasoning, however the gross estimations seem about right to me, and seem conservative compared to some of the estimates out there on the net (some claim results as high as 40%-50% more MPG on older model Honda Civics; exagerated? most probably. Totally fake? I doubt it.) Now does pulling 8A @ 12V from the alternator result in more than a 4% increase in gasoline consumption?
As Guillem pointed out, the flaw is that it will take you more energy to manufacture the hydrogen, than the hydrogen will give back to you. It's that simple. I'm not disputing that adding high concentrations of hydrogen and oxygen into the air-fuel mix will boost power. I'm only saying that that boost in power is still less than what it would take to make the hydrogen/air mixture from water.
Just a wild guess, but I would also be surprised if the typical automotive alternator was more than 50-60% efficient, with the belt drive losses, etc.
Not too long ago I saw a piece on a mainstream news program about one of these systems on a car. Everyone was just delighted about what was happening, but not too much was mentioned about various conversion efficiencies! Back in the early '70s my school, Oklahoma State University, was doing a lot with H2 burning engines. Then, as now, H2 advocates seem to think H2 is just out there for the taking with no production cost (just like the ethanol whackos)! For cars, all the focus seemed to be on how to easily store enough on a vehicle to be worthwhile.
In Spain, a long time ago, Taxis and Buses were powered by Butane gas in orange tanks. But since it was actually cheaper than regular diesel, big oil companies 'convinced' the governement to forbid this kind of cars because they were 'dangerous' and they may 'explode' if they were involved into a car crash.
But maybe what would happen is that this would 'explode' their benefits.
I will bet that if H2 succeeds, that prohibition will be removed, since H2 will be 'less prone to explosions', while H2 + O are in fact the most violent chemical reaction known. But this will be interesting for their busines, and the real danger will be for they earnings.
IMHO, the only place where I would use H2 will be to the application told before, since I'm not aware of any battery technology capable to absorb electrical charge at the pace that braking will produce it, while probably electrolysis will do. And then, use it at a fuell cell, that can be as high as 70% in efficiency and electric motors can be as high as 85 - 95% or even more, compared with 25 - 30% for a normal combustion engine.
Of course, that could be a nice KERS project for F1 applications, so maybe we can apply to Lotus or Renault in order to develop it ;)
Here's some more data I just came by. Chemical energy potential for gasoline is 45.8 MJ/kg, for hydrogen 142MJ/kg. So essentially replacing part of the gasoline in an engine's fuel mix with hydrogen should yield about 3x the raw energy for replaced amount.
I also find your statement "Brown's gas contains H2/O2 in a 2:1 ratio. This means that one pound of gas contains roughly 0.6 pound of hydrogen." to be off base. Your gas should contain a 2:1 molar (molecular) ratio, but oxygen has a much higher molecular mass then hydrogen. Doing the math in my head suggests that this is going to make your gas about 11% hydrogen by mass. (Edit: it's actually 22%. Forgot to account for hydrogen being diatomic).
UNiXWHoRe wrote:Here's some more data I just came by. Chemical energy potential for gasoline is 45.8 MJ/kg, for hydrogen 142MJ/kg. So essentially replacing part of the gasoline in an engine's fuel mix with hydrogen should yield about 3x the raw energy for replaced amount.
You should be looking at energy per unit volume of the fuel/air mixture, not energy per unit mass of the fuel.I also find your statement "Brown's gas contains H2/O2 in a 2:1 ratio. This means that one pound of gas contains roughly 0.6 pound of hydrogen." to be off base. Your gas should contain a 2:1 molar (molecular) ratio, but oxygen has a much higher molecular mass then hydrogen. Doing the math in my head suggests that this is going to make your gas about 11% hydrogen by mass.
Indeed you are right... :) There are so many unknown variables however that I think the maths will wait in favor of observations from now on...
Indeed you are right... :) There are so many unknown variables however that I think the maths will wait in favor of observations from now on...
Will your testing be done with a standard, minimally modified engine?
Any small scale testing on a lawnmower engine?
Do you have a dynamometer available to actually make some comparison tests?
If you have or have access to a dynamometer you could always do some initial testing on the engine part of the system. You can easily purchase large tanks of compressed H2 and O2 to use as a controlled gas source while doing some engine measurements.
This approach would give you some hard numbers on engine performance vs gasoline/gas mixtures, and an indication of what the fuel cell would need to produce for a given "gain" in vehicle performance.
This would lead to a better understanding of how much gas needs to be produced, hence electrode surface area, and current flow, etc.
As you do some preliminary energy efficiency calculations do not forget that you may need a compressor, (or two), to store the gas generated. Unless you are using some clever venturi system you have to inject your H2/02 gas into the engine at the desired rate. Doesn't the fuel cell will generate the gas at ambient pressure?
I haven't done my homework and read any other forums/sites on this, so this may be a dumb question...
Do you feed in the H2 and O2 in a fixed 2:1 ratio? Do you "add" extra O2 to augment the gasoline combustion?
The only reason to use DC for the electrolysis is to separate the H2 and the O2, each being formed at their respective electrodes. AC will still generate gas, but now you have both formed at both electrodes, (still in a 2:1 ratio). This would simplify storage and injection, but what a combination of gases to work with. (Link to Hindenburg goes here: :) ).
Just kicking some ideas around, on a topic in which I have experience of minus 3 on a zero to + 10 scale.
JC
If you are not tied to that 12V source, then you might consider PC power supply. It can provide 5V (if I remember well up to 35A on mine) and 3V3. You can get 2V5 if 5V is split on 2 tanks. Some quick info is here: www.hardwaresecrets.com/article/....
Doc, just some thoughts: store H2 and O2 together is quite dangerous. A little spark, perhaps by some electrostatic discharge, and you have a nice fireworks.
I do NOT plan on storing any significant quantity of gas, it will be produced in the electrolysis tank and immediately used. To keep a constant flow rate there will be a one-way passive relief valve on the tank before the bubbler. Depending on the final container choice, anywhere from 22 to 50 PSI. Apparently a moderate pressure also increases the speed of reaction, but I do not have any hard data on this, so I will have to play around, hopefully without blowing myself up. The switcher I am using also has a programmable current limiter, which I will interface to a little AVR, whose responsibilities will be to monitor temperature, tank level, and current as well as provide feedback to an in-cab LCD... Maybe even control a little pump and valve on a secondary tank for cell water topping. If I can then tap into the CAN (or whatever protocol they use) link of the onboard computer, I should be able to monitor what I need and adjust gas production in consequence, real time, as well as adjust the air-fuel ratio.
If you are not tied to that 12V source, then you might consider PC power supply. It can provide 5V (if I remember well up to 35A on mine) and 3V3. You can get 2V5 if 5V is split on 2 tanks. Some quick info is here: www.hardwaresecrets.com/article/....
For the garage tests this is probably what will feed the switcher, but in the car source will be 12v from alternator/battery.
Will your testing be done with a standard, minimally modified engine?Any small scale testing on a lawnmower engine?
Do you have a dynamometer available to actually make some comparison tests?
If you have or have access to a dynamometer you could always do some initial testing on the engine part of the system. You can easily purchase large tanks of compressed H2 and O2 to use as a controlled gas source while doing some engine measurements.
Unfortunately there is no dynamometer in the garage, but I plan on building a poor man's version by monitoring engine rev speed at a higher than normal idle, along with any other useful parameters I can pull from the engine's sensors via an AVR. I.E. Set idle to 2000 RPM with stock settings and no gas, then introduce the gas and fuel mix changes and check the output. Eventually when I think I am satisfied with the result I might rent some time at a performance garage somewhere to fine tune it.
The tank and electrodes is so simple in terms of engineering i might just build a couple different sized ones, experiment with different plate designs.
As you do some preliminary energy efficiency calculations do not forget that you may need a compressor, (or two), to store the gas generated. Unless you are using some clever venturi system you have to inject your H2/02 gas into the engine at the desired rate. Doesn't the fuel cell will generate the gas at ambient pressure?
The injection itself will be as simple as feeding the gases via the engine's air intake.
Do you feed in the H2 and O2 in a fixed 2:1 ratio? Do you "add" extra O2 to augment the gasoline combustion?
Fixed 2:1 molar ratio yes, as this is how it is produced in the first place. No I do not plan on supplementing the system with extra O2.
The only reason to use DC for the electrolysis is to separate the H2 and the O2, each being formed at their respective electrodes. AC will still generate gas, but now you have both formed at both electrodes, (still in a 2:1 ratio). This would simplify storage and injection, but what a combination of gases to work with. (Link to Hindenburg goes here: :) ).
DC is what is available in the vehicle without hassle. Besides, what about the dead time of AC (when potential falls below the 1.23V reaction)? It would seem to me using a 3.3v, AC will electrolyse half as fast as DC... ;)
Just kicking some ideas around, on a topic in which I have experience of minus 3 on a zero to + 10 scale.JC
I am actually [pleasantly] surprised so many of you have so much interest on the subject. :)
Please do post any definitive results you get from testing this device. I am deeply suspicious that all such devices are inherently inefficient and will ultimately reduce increase your fuel consumption, but I am always open to repeatable and verifiable data indicating otherwise. I have wracked my brain for years trying to come up with a simple way to use solar energy to produce H2, then use that as a fuel supplement. I have always concluded that the energy to compress the H2, plus the added dead weight of the H2 tanks (not to speak of the huge hassle) cancel any benefit. It is better to sell the solar energy to the electric utility and use an electric car powered by batteries. In any event, post your results.
I am deeply suspicious that all such devices are inherently inefficient and will ultimately reduce increase your fuel consumption,
Yep, between the 50-80% efficiency of making Brown's gas from water, and the poor efficiency of automobile alternators, total system performance has nowhere to go but down. Also, you only get about .05L of Brown's gas per KWH of electricity to the cell! I think where you are even *extra* screwed is having the constant high load on the alternator whether you are moving or stopped.
Now if you took the Brown's gas your solar panels generated, and stored it at low pressure in large plastic bags (in your back seat or trunk) you could meter it to your engine and maybe have something 8-) If it accidentally ignites it will not explode, but implode causing a vacuum inside your vehicle.
50-80% efficiency of making Brown's gas
I think I can inprove on that.
8-)
edit: Actually, you have given me a green idea. I should just attach one end of a hose to our new puppy and the other end to my truck's air intake. Everybody wins!
