Originally Posted by Stan_R80/7
Regulators don't typically fail the way you describe. I expect the hose has a leak at a fitting or the torch leaks but the regulator back has a vent that can be tested. Some soapy water makes a good leak detector solution. If you aren't familiar or have not recently read-up on acetylene-oxygen safety, here is a good link: http://www.esabna.com/euweb/oxy_handbook/589oxy1_1.htm
The hose connection wasn't leaking. I always test the hell out of everything with soapy water after changing a tank. It was leaking out the vent holes around the regulator knob. When I couldn't get a reading with soapy water on the fittings, I turned off the main valve, let most of the pressure bleed off, then explored with a lighter. I got some small flames out of the vent holes. Before anyone freaks out and tells me how the tank could have exploded, consider that the main valve was shut off & any O2 had been purged out of the hose. Still, I had an extinguisher handy. Better to find the source of the leak in a relatively safe manner than to risk having my entire house explode.
I'm told that if you forget to back off the regulator and turn on the main valve, the sudden surge can damage the diaphragm. This particular regulator was a Taiwanese Victor knock off that was 25 years old. The O2 companion failed in a similar manner 10 years ago.
Back in welding school, our instructor had the bright idea to have a timed competition in swapping tanks. You had to remove the old tanks, put new ones in place, and light a torch. One torch lighting ignited a small leak from a acetylene regulator. While the students leapt out of their seats and started for the door, the instructor calmly leaned over & blew out the flame.