Sciencemadness Discussion Board - Construction of oxyhydrogen blowpipe

Construction of oxyhydrogen blowpipe

darkflame89 - 9-1-2005 at 02:25

Is it possible to make a simple oxyhydrogen blowpipe? The hydrogen and oxygen is to be supplied by electrolysis, but the trouble is to get it to burn safely?
I need a little help here, the inthernet does not seem to document much about it, only that it is invented to melt metals( 2000degrees flame) and that it is used to produce lime light in the ancient times.
The inventor used 2 sort of chambers to be filled with hydrogen and oxygen. Here I think that is not neccessary, electrolysis will do.

Anyone has an idea how to build one?

Cyrus - 9-1-2005 at 13:08

I've thought of building an oxyhydrogen flame torch too.

I wonder if 15 amps at 120 volts will give a nice sized flame?

Lets see- here's a quote from
http://www.h2cars.biz/artman/publish/article_467.shtml

~

The energy required to produce hydrogen by electrolysis is about 32.9 kW-hr/kg. A kilogram is about 2.2 lb. For 1 mole (2 g) of hydrogen the energy is about 0.0660 kW-hr / mole. For commercial electrolysis systems that operate at about 1 A/cm2, a voltage of 1.75 V is required. This translates into about 46.8 kW-hr / kg, which corresponds to an energy efficiency of 70%. Lowering the voltage for electrolysis, which will increase the energy efficiency of the process, is an important area for research. Intense research persists to better understand ways to improve these hydrogen production methods.

~

So at 66 watts-hour/mol, and if we get at most about 1700 watts of power from the oulet, that means every hour, about 20 moles of H (H2?) will be produced.
Add about 10 moles of O, and I get 15 mols of H2 and O2. Maybe you get something different. Anyways, that's about 300 L of gas per hour, or 5 liters per minute.

Is that enough gas for a useful flame?

Now I probably unbelievably botched that, but anyways, you get the point, I'm not sure you'll get enough gas easily, and storing up gas for later use in homemade tanks would be slightly unnerving.

Of course you don't want the H2 and O2 to mix before they really need to, unless your flame is guaranteed to move faster than the burning rate of H2 and O2. That would be a risky assumption.

I'm assuming a standard transformer and dc rectifier would work, but it would be quite a rectifier!

I have no clue what to use for affordable anodes/cathodes. Anyone?

chloric1 - 9-1-2005 at 13:34

nickel anode/ stainless steel(304 or 316) cathode 30% KOH electrolyte.

Mr. Wizard - 9-1-2005 at 13:49

Do a search for "water welder" or Brown's Gas. There are several overpriced systems out there that burn the premixed (!!!!!), generated gas, sometimes enriching it by bubbling it through methyl alcohol, and using a hypodermic needle as a torch tip. The manufacturers claim it is even hotter than the stochiometric mix, alluding to some atomic hydrogen floating around in the mixture. It is supposed to weld platinum. They used KOH and water as an electrolyte. It looked easy enough to build, maybe even using an old PC power supply.

Cyrus - 9-1-2005 at 16:41

Thanks for the info. Methinks that premixing H2 and O2 and then lighting one end sounds dangerous, but if the Official people do it, then it must be ok.

I don't think a comp. power supply is going to cut it- they put out, what, 35 amps or so? At 1.5 volts, to get 5L of gas per hour, we're talking about 1100 amps!

The_Davster - 9-1-2005 at 17:29

http://www.dangerouslaboratories.org/hydrogen1.html
The amps would have to be upped or the gas somehow pressurized if this was to be used as a cutting/wlding torch of some kind.

[Edited on 10-1-2005 by rogue chemist]

axehandle - 9-1-2005 at 19:03

Plans for a 2H2O2 generator I found while researching the subject, looks a bit complicated though:
http://jlnlabs.imars.com/bingofuel/bgen/

Btw, does anyone know of a catalyst that will make an H2/O2 mix ignite spontaneously? Would a very small Pt fragment do it?

darkflame89 - 10-1-2005 at 03:02

Seriously, this looks complicated..I was thinking that the products of electroysis being used straight away, by going thorugh a sort of compressor? I don't intend to store this mix, its dangerous. Another problem is to have enough current to generate that gas mixture at a high enough rate.

To make H2 O2 ignite spontaneously? I was thinking of using a bit of chlorine(mad idea)..You know, by dissolving just enough sodium chloride to saturate the solution, you could get a mix of H2 O2 and Cl2. UV detonates the mixture spontaneously.

Edit: http://www.cci.ethz.ch/mainpic.html?expnum=23&ismovie=1&...
Look at this, although this has nothing to do with the blowpipe, but the way the hydrogen is generatee and ignited is interesting.. The flame seems to be floating.. Someone needs to translate the page.

[Edited on 10-1-2005 by darkflame89]

mick - 10-1-2005 at 13:19

Btw, does anyone know of a catalyst that will make an H2/O2 mix ignite spontaneously? Would a very small Pt fragment do it?

Adding extra 5% Pd on carbon or 5% Pt on carbon to a reduction mixture saturated with hydrogen and organic solvent will usually ignite. You have to purge the kit with N2 or Ar before adding extra catalyst. I think the activity depends on the surface area of the metal so small bits of the metal might not work or it might need pre-heating. I seem to remember methanol or something similar and air can get Pt wire up to glowing heat. There used to be a cigarette lighter that worked by sucking air through it and the wire coil glowed and lite the cigarette.
mickl
mick

NaOH or KOH ?

axehandle - 18-1-2005 at 13:33

KOH as the electrolyte and Ni electrodes seem to be the textbook method for 2H2O2 electrolysis.

I'm curious as to if NaOH (being very much more available than KOH) could substitute for the KOH, and if so, what caveats apply? Is it only the anode that needs to be Ni? Cathode as well? If not both, what can be used for cathode? All questions, no answers...

I'm going to build a small 600W 2H2O2 generator, and have found a source for nickel, but I'd much rather use NaOH than KOH.

I've tried googling for answers, but all I get is a shitload of links to overpriced commercial systems without any interesting technical documentation whatsoever.

Twospoons - 18-1-2005 at 19:27

The issue of current is easily solved by connecting multiple cells in series. So intead of one cell running 1000A at 1.5V, use 10 cells in series running 100A at 15V total, or 20 cells = 50A / 30V etc...

Plumbing gets a bit more complicated though.

I've seen stainless used for electrodes on some designs - high Ni alloy ?

Mr. Wizard - 18-1-2005 at 20:44

I can't find the site anymore but here is a graphic of a 'water welder' gas generator. Note the step down transformer, diode, electrolytic cell, and one way valve in the plastic line. My apology to whoever owns the picture.

ww1.jpg - 65kB

Mr. Wizard - 18-1-2005 at 20:46

And here is the flame, at the end of a hypodermic needle. Also not my picture, I'd give credit if I could.

photo by:
I found the address in the user groups.
rec.crafts.metalworking
http://www.speff.com/ww1.jpg

[Edited on 19-1-2005 by Mr. Wizard]

wwflame.jpg - 44kB

uber luminal - 18-1-2005 at 22:05

So we can build a cell, or several cells that have seperators over the anode and cathode to gather the 2 gasses independently.

Thats the easy part. The hard part is this. You wont be able to mix the two gasses and have a continuous flame unless you have a forced feed, and by that I mean you need to maintain a certain flow. The flow needed will depend on the oriface where the gas will be burned, and the speed that the flame moves. If your flow is too low, the flame will run back into a turbulant place in your pipe or as far back as you mixing chamber (called a flashback). the flame will continue to burn up the gases thus far back and either choke eventualy, or create a well and run back into your H2 source with O2 running right behind it, spreading your equipment around the room with a nice orange ball of energy.

If your flow is too high, the flame will mip, (dance at the end of the gas with a gap between it and the oriface) or, it wont even light.

notes about hydrogen as a heating gas. when combusted with the right proportion of O2 (2H2 +O2) it could in theory produce an inner envelope that has a temp of about 5300 F.(this is the gross, not the net) However... Hydrogen has a very low density, so its heating value for one cubic foot is very low compared to other gas fuels that are made of larger molecules. H2 also has a higher flame speed than most other fuel gasses, so you must burn MORE H2 to sustain your flame than any other gas fuel.

The smaller the oriface the better when it comes to H2/O2 otherwise you will have a hell of a time forcing hydrogen into your pipe at the rate needed to sustain the flame.

I could go on and on about the different things you would need, (check valve, flashback arrestors, some way to regulate your endflows for mixing, shut off valves, some way to force feed the gasses (put them under pressure), etc) The most important being a way to exactly regulate the mixing of H2 and O2. you wont get your 3000 F flame if you dont get it exactly right. the changes are quite drastic when you make a carburizing flame vs neutral vs oxidizing flames, and I imagine the difference in the gas mixture of H2 and O2 could be very small to produce the different mixtures. This may be the reason a "professional" product would have had the gasses premixed since they are allready in the right proportion from 2H20 -> 2H2+O2, and since you are reacting the opposite, hell thats great... and really hazardous.
but this method would not allow you to do any cutting, unless you used 2.89x of compressed air along with it, but then you get nasty nitrates and CO2 to deal with.

My thought would be... this is a cool idea, I mean having a cheap way to produce a hot flame that you could just bleed when your done. No storing hazards or having to pay fees etc. But I really think that even if you could make a system work, it would not be safe. at all. you would not save any money because you could weld 22-26 gauge steel with it.(haha)
You definatly would not be able to do much cutting with something like this, unless you were able to pump pure O2 down a 2nd oriface that wouldnt react with the 1st oriface (so the 2nd path of O2 can go to react with the metal rather than the H2).

If you want something cheaper... try getting a butane, propane or MAPP canister. its cheap, its easy, its a hell of a lot safer. If you have more money, *rent*(buying them costs a lot of cash. I think mine were $200 a pop) some gas cylinders and buy some acetylene and oxygen. Oxy/Acet is an industry proven method of reaching 5600 F, and its been around for... a looong time.

on a last note, I have seen hydrogen gas cylinders hooked up to oxygen cylinders with oxy/acet torch equipment used... it does work, its just not very practical without pressurized cylinders.

Oxydro - 23-1-2005 at 13:18

I've tried something like this before, but I couldn't get the flame to sustain all that well. This site is the best resource I've found so far, even if the guy is into overunity and all that crap

.

If you try it... try filling a 1L pop bottle with the mix and igniting it (remotely) You will find only fragments of the bottle. For something really impressive, try the 5-gallon watercooler bottles. Althoug you need a lot of gas to do anything usefull, its explosive power if mistreated is rather scary!

Make sure, if using mixed gas, to include a bubbler in your system (use water to isolate the storage from the flame). Make the space above the bubbler as small as possible.... it only takes a few CCs to make an earsplitting bang and send the bubbler flying (the pop bottle I was using would compress and then bounce off my desk).

My brother tells me that oxy-hydrogen flames are used for welding of small wires eg platinum electrodes -- he used to design paper-plant sensors for Honeywell, and he contracted a jewler to use such a torch for delicate assemblies.

rift valley - 24-1-2005 at 20:28

Wow thanks for the link, a lot of interesting stuff on that website, I have some questions though. Here is a picture of the cell he made it in corporates 80 6"x6" #316 stainless steel plates (that couldn't have been cheap). The plates are spaced a few mm apart with noncunducting plastic. The electrodes are attatched to each side, so there are 78 metal plates in there with no direct contact with an electrical source. Is the reason for the extra plates to improve the conductivity of the cell? For some reason it is my instinct that one electrode should be in direct contact with 40 of the plates and the other electrode in contact with the other 40, is it wrond to think this? Her is the picture I have been referring too.

Before going to that link this thread was just a novelty, but now I am genuinely interested in pursuing this.

Thanks,
Rift Valley

P.S. Bubbleing the gases through water appears to be a fail safe way to prevent a flashback, or at least containing it to post-bubbler plumbing.

Chris The Great - 1-2-2005 at 03:10

I would think that each plate would get a voltage difference of a few volts between each one, and so when when a higher voltage is applied it is able to be done at a much lower current, because the cell will be like 40 2V 10A cells, so the total current is 400A, even though the input is only 80V at 10A. That way the current is used as efficiently as possible.

At least that's what I would think happens. If so you could run the thing off rectified wall socket power, at 160VDC average. If you divided it to have 2V between each plate, and had an input of 15A, then you would have an effective current of 1200A, without having to build a transformer that can handle 1200A (not an easy task at all).

Assuming I'm right with my idea that is, but it seems like it should work.

Thread reboot

watson.fawkes - 14-12-2010 at 08:53

Based on pictures of an oxyhydrogen torch in this thread, I decided to do a little research on them. Here are some results.

Rio Grande (jeweler's supplier) sells what they call a "water torch" (category link), the Hydroflux Welder by Okai. Rio Grande sells NaOH for electrolyte; Okai sells KOH. The Rio Grande "Water Torch Flux Dry Mix" is boric acid. It's dissolved in methanol to make a dilute solution. Electrolysis gas is bubbled through this; it picks it up as vapor. The effect is to make the flame slightly acidic and slightly reducing, both good for jewelry work. There's a gas line filter for the unit with a sintered bronze filter tube. It's available as a fuel filter from JCWhitney. I'm sure this filter also acts as a flame arrestor; it seems like a generally good idea to use these sintered metal filters for this purpose.

The tips on this burner are simple Luer Lock hypodermic needles. McMaster-Carr sells them under the category "dispensing needles". In this same category, though, they also sell heavier-wall needles that might work better mechanically. They've certainly got more heat capacity and thus would cool a flame front better. McMaster-Carr also sells Luer Lock couplings to put on the torch head. @DougTheMapper: You can get easily switchable needles with this system, rather than hacking it together with a barb fitting and heat-shrink.

From reading a discussion board something (sorry, reference lost), the Hydroflux operates at 3 psi gas pressure with some kind of manostat that controls average electrolysis current. There's another brand of jeweler's torch with adjustable output.

DougTheMapper - 14-12-2010 at 14:27

I have decided to repost the start of my BYMC thread here since the topic at hand is a lot more relevant than the site where it was originally posted. Furthermore, I feel like I can contribute useful material this way without constantly referring to another site. There was a bit of Q and A in the original thread that I'm not going to transfer, but I can always re-answer them here. Anyway, here's the post:

Hey all:

I just wanted to show you some photos of my oxyhydrogen torch project.

First off is the power supply. This is just my rewound microwave oven transformer to which a trio of parallel alternator diodes have been added.
As you can see, the diodes have been bolted to a large heatsink because they dissipate quite a bit of power while rectifying 15VAC to half-wave DC at 25A.

Here is a picture taken during the construction of the cell itself. All hardware is either 304-series stainless or nylon. 316L could be used for longer-lasting plates, but that would make this cell significantly more than $10. Stainless steel is pretty hard to drill though because its chromium, nickel, and manganese content make it very hard. I used a bit sharpener several times.

Here is the final assembly. The margarita bucket is the main cell and the two stainless steel rod terminals can be seen exiting the top. The vinyl tubing leads from the cell to the bubbler which acts as a flashback arrestor since the oxyhydrogen mix is extremely explosive. The blue stopper is just some rubber piece that happened to fit in the vent hole and should prevent total cell destruction should an explosion occur. It is also used in conjunction with the tap at the bottom to swap the electrolyte. The bubbler is contained in a second bucket to catch the water released if it explodes.

Here is a view into the cell after about a half hour of continuous operation at intermittent intervals over the course of an afternoon. The iron oxide flocs in the electrolyte are expected and are a result of low-quality stainless steel. The plates show almost no stains or signs of wear and should last for many hours of operation before they must be acid-etched to clean them. The electrolyte is a solution of NaHCO3 in water and has been adjusted in concentration so the cell draws 25A when fully warmed.

The following photo shows the gas generator hooked up and in operation (note the bubbles present in the bubbler device and the distortion of the cell lid from gas pressure).

Unfortunately, the following photograph is a bit blurry. It shows supply current and voltage drop across the cell. Admittedly, the battery in my voltmeter is dying, so it reads a bit low. The cold cell is drawing 22.7A and dropping about 5V for a total power of circa 113W.

This is the torch itself consisting of some segments of half-inch galvanized iron pipe and various fittings and valves. The interior of the tube is packed with very fine (#0000) steel wool. When a flashback occurs, the wool draws enough of the heat from the flame front to arrest it and prevent the bubbler from exploding. This has been isolated and tested about 50 times with no breakthrough of the flame wave so I deem it a suitable safety device.

The following is a picture of the flame in a darkened room.

It may not look like anything significant, but you might be surprised. The incredible thing about this flame is that the two gases burn so fast that a visible flame is only 8mm long - however, the hot exhaust of the flame carries much farther and will ignite paper over 4cm away, demonstrated here:

It's also easy to underestimate the heat this flame generates. Here, it has been trained on a 90/10 silica/bentonite refractory coupon. As you can see, this torch has no problem vitrifying my 3000+°F refractory, something impossible to do with a even a propane torch. For reference, I'm burning hydrogen in oxygen, which has an adiabatic (theoretical maximum) flame temperature of 5792°F. This tops an air/propane torch which maxes out at 3596°F and even rivals oxyacetylene at only ~400° cooler. Neat!

Anyway, the diodes tend to get hot fairly quickly so I have found a set of 5 power MOSFETs on a heat sink which should help to alleviate this problem when configured to replace the diodes. I must also mention that the current torch made desoldering these MOSFETs and the heatsink an absolute dream.

Well, that's about the end of the latest lab developments. I hope you enjoyed!

-Doug

EDIT: Thanks for the info about the needle couplers. I'm still waiting for some money to open up for more robust rectifiers. I'm spending nearly $300 on glass and reagents for the holidays so I'm probably not going to get this project rolling until I actually have a use for it other than burning the crap out of things.

[Edited on 14-12-2010 by DougTheMapper]

crazedguy - 14-12-2010 at 15:37

This seems like a really good idea and very cool. Maybe try not lighting it fill a balloon tape fuse to balloon = KaBooom.

AJKOER - 29-5-2011 at 16:48

An idea: To ignite H2 gas place it in contact with Cl2O (Dichlorine Mono-oxide) gas. Ignition to a mild explosion should result.

To prepare Cl2O: Pass a steady stream of air over concentrated HClO (Hypochlorous Acid) that has been saturated with Ca(NO3)2. This works as the concentrated acid easily releases Cl2O (the anhydride of HClO) and Ca(NO3)2 is a drying agent.

Other methods: Pass Cl2 over heated NaOH. Or, Cl2, an inert gas and steam over heated Sodium carbonate.

Note: Due to it unstable and explosive nature with shock and light sensitivity, Cl2O should not be stored.

Master Triangle - 24-12-2013 at 23:21

I am considering building an oxyhydrogen torch apparatus and am seeking feedback on the design.

The electrolysis cell would be constructed with about 50-60 plates so that I can use it directly from 120v from a stepdown transformer. I considered building a cell or series of cells that you could apply rectified 240v across directly but I think that a little electrical isolation is in order, what with all this electrolyte and possibility of explosions.

For the people who are still wondering about the stacks of unconnected plates in most cells they basically divide the voltage as the optimal plate-to-plate voltage is somewhere around 2v and having a higher voltage just makes heat. There is a voltage gradient between the two end electrodes and just putting an unconnected plate in between them will basically divide the cell into two cells of half the voltage in series, the current will remain the same however so gas production is basically doubled.

One of my possible design ideas was an ultra-cheap method where you cut a thin stainless sheet into appropriate plate sizes, apply RTV or liquid gasket around the edges, and sandwich together with some thicker end plates that have screws extending down outside the stack to hold it together. You will have to avoid angled plates, if they are not parallel then there will be greater voltage and hence current flowing in the close area and the rest of the plate area will be doing very little.

The cell voltage could be adjusted using a PWM controller before the rectification stage, some electrolytics would increase the efficiency a bit too. The pressure would be maintained by having a pressure switch that turned on and off the cells to maintain a roughly steady pressure, or even better, a PID controller.

Then have the usual flashback arrestors, sintered metal seems like the best option, I don't think I will put a bubble chamber in. The torch would be a simple hypodermic which can be easily swapped out for various sizes, I will try to incorporate the flashback arrestor into the hand piece to reduce the unprotected volume.

I am wondering about the burner design though, how wide can you make the nozzle before flashback occurs? Does it depend on gas velocity through the nozzle? Can multiple nozzle be used aimed at the same spot? I don't think I will bother with any exotic methods of ignition, a lighter seems to work well for most people.

Also the problem of spud guns using oxyhydrogen, it burns at about 1-2.5km/s producing a dangerously high shockwave and pressure peak, then the pressure drops rapidly because the h2o just condenses. It is hard to get any work out of it because work=force*distance and it is hard to get anything to go very far during the pressure peak. If anyone knows how this could be made to work (haha) better then that would be great.

TheChemiKid - 25-12-2013 at 08:18

Quote: Originally posted by axehandle

KOH as the electrolyte and Ni electrodes seem to be the textbook method for 2H2O2 electrolysis.

I'm curious as to if NaOH (being very much more available than KOH) could substitute for the KOH, and if so, what caveats apply? Is it only the anode that needs to be Ni? Cathode as well? If not both, what can be used for cathode? All questions, no answers...

Sodium Hydroxide works almost as well.

Zyklon-A - 25-12-2013 at 10:19

Might try to build one of these. Is a US nickel pure Ni, Plated Ni, or alloyed Ni?

Metacelsus - 25-12-2013 at 13:52

A US nickel is 25% Ni 75% Cu. There is a thread on extracting the nickel.

https://en.wikipedia.org/wiki/Nickel_(United_States_coin)

Zyklon-A - 25-12-2013 at 15:49

So it's an alloy then? Too bad it's not Ni plated.... But stainless steel works as an electrode right?

jock88 - 25-12-2013 at 15:59

If you go to your local welding supply shop they may give you a sample of Ni welding rods. These are (I presume) pure Ni. I obtaineda bout fifty 3 mm by one 10 inch long rods this way.
You can of course purchase

Zyklon-A - 25-12-2013 at 16:15

So Ni is really needed then?

jock88 - 25-12-2013 at 19:04

I don't know if Ni is actually needed. Would not carbon rods do too. They are available as gouging rods.
The cathode can be made from just about any metal as it will not corrode. The anode must be non corrodable in the KOH solution. If Ni will do I would think that other relatively inert metals would do as well, stainless steel being most easily available.
Valve metals (Ti etc) will not do as they will not conduct electricity when used as an anode. OK for a cathode.
Would a great big piece of carbon fiber cloth (if you have any) do the job. Is the stuff conductive?
The drum of a washing machine from the dumpster sounds good to me.

Zyklon-A - 25-12-2013 at 19:13

KOH is just the electrolyte right, can't you use something else, NaCl, ect.

Master Triangle - 14-1-2014 at 18:39

Using NaCl will electrolyse out the chloride as chlorine gas which may damage your device or autoignite the oxyhydrogen mixture (although I have not seen this listed as a warning anywhere?). Since it is aqueous any group 1 metal like Na or K will do for a cation but the anion should be something like hydroxide. I believe carbonates will also work fine?

Zyklon-A - 28-1-2014 at 15:46

I heard that O2 lessens the likelihood of Cl2-H2 to autoignite, could be wrong though.

BromicAcid - 28-1-2014 at 16:51

Was looking up information on buying one of these the other day, I was wondering how hot they get since the electrolysis will not hit theoretical output and found this video:

<iframe sandbox width="560" height="315" src="//www.youtube.com/embed/GaOizadevuM" frameborder="0" allowfullscreen></iframe>

I found it of course because I was searching for HHO after Bfesser's reprimand of another member for crackpot science. Back to the video though, they sell these small generators for 'acrylic polishing' on eBay and this was the only video I could find on one in action. Of course the brain trust that shot the video decides to test the mettle of the flame against the might of thin plastic, wood, and a digital thermometer (for food it looks like) that goes up to only 300C. It's sad really.

Zyklon-A - 28-1-2014 at 17:10

Lol, the guy says "The flame is not really burning, it's just recomposing into water".

He also says "you can use the flame to heat your house".
No but really, it looks like a very good replacement for a propane torch, I would buy it if I could.
Don't use it to heat your house though.

[Edited on 29-1-2014 by Zyklonb]

The garageguy - 20-6-2017 at 13:02

This is my first post, so wish me luck

I don't know much about this subject but have built many many hydrogen cells and eventually settled on a dry cell, a 6-7 plate dry cell running at 12v and a current density of less than an amp per cubic inch will have a reasonably high gas output and be very very efficient in comparison to a wet cell.

This is a link to a video about a cell and torch I made about a year ago.
https://www.youtube.com/watch?v=vYz3U8GmE-0