There are couple of things I would like to add here after reading some answers.
1. Blowing into the supposedly empty cylinder with your mouth is a quick way to ensure the cylinder is empty. If you can blow air in, it's empty, if you can't it's blocked, simple as that.
However, I learned the hard way not to do so, after spitting out a mouthful of aluminium-oxide. If you empty a full, wet and aluminium-oxide laden cylinder quickly, it will usually drain most of it, stop, drain some more, stop, drain some more, and so on. I have been impatient and blew into a cylinder prematurely before it was completely empty.
The best way is to use low pressure compressed air to blow a little air into the cylinder, achieving exactly the same without having your mouth there.
2. Weighing cylinders is, as pointed out above, inadequate - always. Weights are not always stamped onto cylinders and batches vary by quite a bit when manufactured. There is also no real way to check if maybe another 30bar are left in there while taking slight variations in cylinder weight and scale imperfections into account.
3.
Removing the burst disc is not advisable. Cracking it ever so slightly open however can be a good way to empty a cylinder.
4. I do not like the idea of just putting a wrench to it and trying to open it. While it is true that is exceedingly hard to do so while pressurized, it is by no means impossible. Countless accidents have happened, simply because people do not know how much force they are expected to exert to remove a valve. I know one person who is very lucky to have survived opening a full aluminium cylinder by force, but he still carries a souvenir of aluminium lodged just underneath his eye.
For an technician who does this day in and day out, it is trivial to see when something is amiss, like the cylinder full or extremely corroded. For the layperson who does this the first time, it may not be apparent and the thinking often is:
Well that cylinder had to hold a lot of pressure, so it makes sense that it was tightened so much! Of course this thinking is flawed, but I do not fault anyone for thinking this way, it somewhat seems
logical at first.
Furthermore there is a continuous line between how full the cylinder is and how hard it is to unscrew the valve. The same continuous line exists between how corroded the threads are and how hard it is to unscrew the valve.
Just trying to unscrew the valve does not let me discern if I made a mistake and the cylinder is still full, or just extremely corroded.
4. Shooting at cylinder... I occasionally forget from which part of the world most members here come, but that certainly was a great reminder!
The only practical way that I know of to tell if a cylinder is empty is to blow into it with low pressure air. As mentioned above, preferably from LP source other than your lungs. A old LP hose from a regulator hooked up to a first stage is excellent for this. One will hear a distinct sound of the air flowing into the cylinder and then immediately out of it after removing the LP hose.
If the cylinder is not empty after the above test, I would always start at the "knob" side.
- Is just the knob broken? Take it off and see if you can twist the spindle with an appropriate tool. This scenario is extremely common.
- Is the stem broken? See if it's a valve that will drain when opening the packing nut. Some Z-Valves, like Lavo's empty this way. While doing so, check that the spindle does
not move and just
crack the packing nut open a
little. If it does not drain via the packing nut, see if the seat slot is still in tact. A crude tool resembling the end of a stem can be fashioned out of an old bolt and that almost always let's you twist the seat.
- Is the seat slot sheared off? That sucks, hopefully this valve has burst disc. Ever so
slightly crack it open and then
stop. Let it drain, even if it takes time.
- The last option is to drill into the valve. Quoting from another thread:
...thick gloves, thick jacket, face-shield, protective goggles, ear muffs, etc. are all necessary. You aim as low into the air-path as you reasonably can - Think about the continuation of the dip tube, that's what you are aiming for. The hole gets drilled with a tiny drill, like 1.5mm. Once you get "through", the metal shavings from drilling usually fly back and lodge themselves somewhere in the face-shield or jacket.
While doing any of the above options, not counting the broken knob scenario, it is always a good idea to fixate the cylinder. Wearing safety googles is not a bad idea either. I got lucky once and hit on the forehead with enough force to cause a tiny cut by a completely dried up and crumbling O-Ring from a packing nut. If it had hit my eye, it may have caused some damage - or not. Either way I did were safety glasses from then on out.