What's new
What's new

Did I just almost die? (oxy-acetylene)

birdus

Plastic
Joined
Nov 24, 2013
Location
United States
I was just out in the garage welding some 4130 with my oxy-acetylene setup. I'm a bit of a neophyte, but this was unexpected. I was getting some popping from the tip, but that wasn't the first time. The tip had gone out a couple times, but I just turned off the oxygen, stuck the tip next to the hot metal, got re-lit, added the oxygen back in, and kept going. No biggie.

The last time the torch popped, I heard a sound come from the tanks. I glanced over, then glanced back at the torch (which was still running), then glanced back at the tanks. I'm guessing it's a good thing that I could see the acetylene tank ON FIRE through my dark mask. I flipped up the mask to see if I really saw what I thought I had seen...and I had.

I killed the torch, then walked briskly over to the tanks, reached into the f*&#ing fire and turned off the acetylene tank. The fire continued to burn (presumably the gas still in the hose) as I tried to blow out the fire. I grabbed the valve just to make sure I had turned off the tank. I had. I blew on the flame some more, expecting my head to get blown off at any second. Finally, the flame went out. I turned off the oxygen. Then I bled the hoses.

Then, I decided I would go into the house because I still didn't want to die in a gruesome explosion. I figured I would find out what happened so that I wouldn't die finishing the welding later on.

Two things. First, the flame was coming from the joint where the regulator attaches to the tank. It would seem I didn't tighten that enough. Second, isn't the blowback preventer (or whatever it's called) supposed to prevent this kind of thing? In the photo, you can see where the flames were (it's black) and you can see the thing that I thought was supposed to prevent this.

WTF?????

p2854230759-6.jpg
 
Blow back preventer will NOT do anything for a external leak at any point!! They only work for the internal passage back flow. Not a bad habit to check your connections with a soapy water spray before use especial after tank change outs. All it took was a random spark close to the tank with a small leak there.
 
Some people should not be allowed to play with:
1. Torches.
2. Plasma cutters.
3. Welders.
4. Hydraulics.
5. Guns.

And then you wrote:
"The tip had gone out a couple times, but I just turned off the oxygen, stuck the tip next to the hot metal, got re-lit, added the oxygen back in, and kept going. No biggie."

Yes "Biggie", stop disrespecting the torch/process.

A small bottle of "Snoop" leak detector to be used every time
you change out a bottle seems to be in order here.

BTW Time for you to fix you location too.
 
Were I you, I'd be playing the lottery tonight and thanking my deity(ies) of choice.

Best get some education/instruction before trying to die again.

You DO know that Acetylene can ( and will ) self ignite at approximately 15 pounds. Right? How much pressure is in the tank...? ( rhetorical ) Go buy that ticket.
 
Were I you, I'd be playing the lottery tonight and thanking my deity(ies) of choice.

Best get some education/instruction before trying to die again.

You DO know that Acetylene can ( and will ) self ignite at approximately 15 pounds. Right? How much pressure is in the tank...? ( rhetorical ) Go buy that ticket.

He might as well save his money, he's used up all his luck for the next decade.
 
Ummm not to be a dick, but there was a lot wrong with what you posted. I don't claim to be an expert, but lighting the torch on hot metal just sounds like a bad idea. That's not very controllable and I could easily see a large cloud of acetylene building up then flashing. They make strikers for a reason.

How can the flash back arrestor do anything about a leak downstream from it? Think about it. Also as I understand it a flashback arrestor is to stop flame from reaching the bottle.

On the acetylene side, you should just crack the bottle open enough to allow flow, but so that you can close it fast. I generally open slowly wait for the pressure to come up on the bottle side of the regulator and then give an extra 1/4 or 1/2 turn.

On the oxygen the valve is design to seal either fully closed or fully open.

I also check for any leaks every-time starting the torch.

And I don't do anything with the torch without a good sized fire extinguisher very close. Dad set the fire on fire when I was about 7, I and remember having to fight the fire with a garden hose and hoe to keep equipment from burning down. Not fun.

Grandfather also got his forearm raked by a cutting torch, even just a short exposure left quite a wound.

O/A needs to be respected.
 
Ummm not to be a dick, but there was a lot wrong with what you posted. I don't claim to be an expert, but lighting the torch on hot metal just sounds like a bad idea. That's not very controllable and I could easily see a large cloud of acetylene building up then flashing. They make strikers for a reason.

How could a large cloud build up when I re-light the torch within a second or two of it going out? How would using a striker improve upon that? I've seen other experienced welders relight it on the metal. Maybe you're right, though. It could be a terrible idea. I just don't understand why it would be.

How can the flash back arrestor do anything about a leak downstream from it? Think about it. Also as I understand it a flashback arrestor is to stop flame from reaching the bottle.

That's what I understand, too, which is why I was confused as to how the fire started. No one here has told me how they think the fire began. I assumed it traveled up the hose. Again, thus my confusion. I was probably eight feet from the bottles. If the fire began because of a large cloud of acetylene coming from that leak, why wasn't there a big boom when it ignited? There's no evidence it was lit from my torch.

On the acetylene side, you should just crack the bottle open enough to allow flow, but so that you can close it fast. I generally open slowly wait for the pressure to come up on the bottle side of the regulator and then give an extra 1/4 or 1/2 turn.

That's exactly what I do.

On the oxygen the valve is design to seal either fully closed or fully open.

That's exactly what I do.

I also check for any leaks every-time starting the torch.

At a minimum, I'll do a better job checking for leaks when hooking up to new tanks.

And I don't do anything with the torch without a good sized fire extinguisher very close. Dad set the fire on fire when I was about 7, I and remember having to fight the fire with a garden hose and hoe to keep equipment from burning down. Not fun.

Good advice about the fire extinguisher. That's the first thing (or one of the first things) that went through my mind when I saw the fire. Of course, I was pretty busy trying not to panic.

Grandfather also got his forearm raked by a cutting torch, even just a short exposure left quite a wound.

O/A needs to be respected.

Thanks for your effort to help me out.

Jay
 
Blow back preventer will NOT do anything for a external leak at any point!! They only work for the internal passage back flow. Not a bad habit to check your connections with a soapy water spray before use especial after tank change outs. All it took was a random spark close to the tank with a small leak there.

I assumed the fire began from internal back flow. That's why I was confused. If a spark started the fire, I have no idea where it would've come from. I appreciate your thought about that having been the cause, though. I'll definitely start checking those connections. Thank you.

Jay
 
How could a large cloud build up when I re-light the torch within a second or two of it going out? How would using a striker improve upon that? I've seen other experienced welders relight it on the metal. Maybe you're right, though. It could be a terrible idea. I just don't understand why it would be.



That's what I understand, too, which is why I was confused as to how the fire started. No one here has told me how they think the fire began. I assumed it traveled up the hose. Again, thus my confusion. I was probably eight feet from the bottles. If the fire began because of a large cloud of acetylene coming from that leak, why wasn't there a big boom when it ignited?

Key there is "experienced", that lets you get away with a lot. Pro truck drivers shift without using a clutch, if you haven't spent enough time in the seat, floating gears might as well be magic.

What I'm thinking of is the possibility were your got fuel gas coming out, and your material may not be hot enough and then you have some external spark, or reaching a hotter bit of material and instead of starting uncontrollably, you get a larger flame. Unlikely I grant you.

But I'd trust a striker over your way for a lot of reasons. One the main factors in causing an accident is skipping mundane steps that for someone unknown reason or unfortunate bit of luck end up biting you in the ass. But mostly cause I don't like it and wouldn't feel comfortable doing it.

I just watched a documentary about an accident in a Titan II missile silo. PTS team used a ratchet instead of torque wrench to install or remove (cant remember which) a large nut in order to fuel the missile. The rachet didn't hold onto the socket and the 8 lb" socket ended up getting drooped about 100' and punching a hole in the side of the missile. Caused a fuel leak and eventually the missile blew up and having a very large nuclear warhead go off in the middle of rural Arkansas was a very serious concern.

Obviously an over exaggeration, but the point still stands.

The thing about a striker is that is a spark, gone very fast, no open flame and very controllable. It's the safest way IMO. But you do you.

Pretty sure flowing gasses generate static electricity by their nature, not much, but that's what I'd assume started it.

And your fire was burning off a leak, the fire was consuming the gas coming out. Had it you not had the fire, and been working in a confined poorly ventilated space, I would assume youd have a LARGE bang as the space filled up with acetylene and eventually reached the torch.

There is much that scares me, but a fire on a stick that is several thousand degrees fueled by pressurized gas gets my undivided respect. To the point that I may look like an idiot and sound like a safety nazi.

Another thing to consider, from the OP is sounded like you were working in a garage. So I'm assuming the house is attached and perhaps family members inside? To me that sounds like a reason to be even more careful.
 
If the flame burns up the hose there should be evidence inside the torch body like some smoke or maybe even burned o-rings. It would also make a nasty howling noise.

The easiest way to see if the connections are leaking is to have the torch valves closed and open the tank until pressure equalizes and then close the tank. With the torch valves still closed the needles on the guages should not move. If you want to be sure leave it foor about 5 minutes and see if needles have moved. If the needles do indicate a loss in pressure THEN get out the leak check/ soapy water if you need to.
Dont worry about re lighting your torch off the hot metal either, nothing is wrong with that.
 
Watched that same documentary a couple months ago. :) Totally insane stuff there.

Detached garage. Wife out of the country. Don't really like the neighbor who was just on the other side of the wall from the tanks, so no problem. I'll fix the leak before the wife gets back. :D

I truly appreciate your time and thoughtful replies.

Jay
 
Is what's connected to your regulator an actual combination flashback arrester and check valve, or is it just a check valve? A check valve only won't prevent a flashback from reaching the tank. It will prevent the back flow of gas, but it won't stop a fire from burning it's way back to the tank.
 
The puzzle, as you suggest, is how the acetylene got ignited back at the leaky tank fitting.

1) I'd check the hose to see if there is any sign at all of a flashback (and replace the hose and regulator innards if there is). Some slight chance this is the cause. Is that just a check valve at the tank or also a check valve and flashback screen?

2) As also suggested, could have been some source of ignition you didn't notice.

3) Over-pressure another slight possibility.

4) ???

You were lucky. Mixtures of air and acetylene are explosive over a wide range. Seems your leaky-fitting-subsequently-ignited scenario somehow avoided that.
 
If the flame burns up the hose there should be evidence inside the torch body like some smoke or maybe even burned o-rings. It would also make a nasty howling noise.

The easiest way to see if the connections are leaking is to have the torch valves closed and open the tank until pressure equalizes and then close the tank. With the torch valves still closed the needles on the guages should not move. If you want to be sure leave it foor about 5 minutes and see if needles have moved. If the needles do indicate a loss in pressure THEN get out the leak check/ soapy water if you need to.
Dont worry about re lighting your torch off the hot metal either, nothing is wrong with that.

Thanks for the reply, Rob. That's some good info. I'm glad to be learning some things.

Jay
 
The torch pops because you are not running gas through it fast enough or you are getting the tip hot. Acetylene/oxygen has one of the fastest propagation rates known. Given the chance it will burn up the tip and ignite the mixture between the mixer and the end of the copper tip. These torches use copper tips for good reason. They need to conduct heat away fast enough that the mixture is cooled below the ignition temperature. A high enough gas flow helps cool the tip. Also, radiant heat from the weld puddle makes life more difficult. That is why it pops when you have the biggest puddle. The noise you heard, probably a sort of whine, was from burning in the tip. It would have melted off shortly. A dirty tip will also cause trouble.

The flame will not burn back in the hose unless you have the oxygen pressure so high that you get backflow. The leak at the tank was leaking all the time and a spark from the puddle ignited it. To prevent leakage at the tank, I cut an O ring groove in the fitting and now get a reliable seal.

Get professional help before you have a real disaster.

Bill
 
Is what's connected to your regulator an actual combination flashback arrester and check valve, or is it just a check valve? A check valve only won't prevent a flashback from reaching the tank. It will prevent the back flow of gas, but it won't stop a fire from burning it's way back to the tank.

Since the tanks made a funny noise (a thud?) and the fire began at the same time as my torch popped, I assumed the flame had traveled up the hose, but it sounds like that couldn't have been the cause of the fire.

Jay

p2854383574-5.jpg
 
The puzzle, as you suggest, is how the acetylene got ignited back at the leaky tank fitting.

1) I'd check the hose to see if there is any sign at all of a flashback (and replace the hose and regulator innards if there is). Some slight chance this is the cause. Is that just a check valve at the tank or also a check valve and flashback screen?

It's the whole ball o' wax. See a photo of it that I posted in answer to another post in this thread. What signs of flashback should I look for in the hose? What are the indicators?

2) As also suggested, could have been some source of ignition you didn't notice.

Unless the flashback arrestor failed (if that's even possible), and I would assume it would be totally clear if the torch had ignited a cloud of acetylene, some spark seems the most likely. Another poster suggested static electricity caused by the flow of gas.

3) Over-pressure another slight possibility.

What do you mean? The tanks are mostly full and I was running the regulators at around 5-6 PSI.

4) ???

You were lucky. Mixtures of air and acetylene are explosive over a wide range. Seems your leaky-fitting-subsequently-ignited scenario somehow avoided that.

Thanks, Pete.

Jay
 
"On the acetylene side, you should just crack the bottle open enough to allow flow, but so that you can close it "

common practice is to open the valves W.A.O. to keep the seals in order. there's
not a lot of tank pressure in acetylene dissolved in acetone . waaay less than
oxygen or argon tanks.

regardless... be careful of any compressed gas.
 








 
Back
Top