Sciencemadness Discussion Board

Pre-mixing lpg/propane with air safety issue.

Panache - 11-8-2011 at 22:31

are there any?


Ok more detail required perhaps, the question relates to the run between your mixing 'chamber' (which takes the compressed air and lpg feeds) and the torch. The reason i'm concerned is i have never seen lpg/air pemixed cylinders, understandably under compression things are less safe, but could i have a long run in the workshop, say 20m or is this unsafe, why is it safe in a torch?

do i make sense?

watson.fawkes - 12-8-2011 at 04:46

Quote: Originally posted by Panache  
Ok more detail required perhaps, the question relates to the run between your mixing 'chamber' (which takes the compressed air and lpg feeds) and the torch. [...] could i have a long run in the workshop, say 20m or is this unsafe, why is it safe in a torch?
The issue is flashback, which is propagation of the flame front backwards into the torch tip, whence it just keeps going. Propagation will continue as far back as it can, either to a flashback arrestor or to the mixing chamger itself. Flashback happens when the gas velocity through the tip is slower than the flame propagation speed. Good luck trying to adjust flow rate at the far end of a 20m run and not have huge flashback problems.

One basic safety principle is to minimize the total amount of stored combustion energy in the mixed gases. The easiest way to do this is to mix fuel and oxygen in the torch body. There's a reason this is the standard technique. The only other standard practice I know of is common for glass working. It uses bench-mounted venturi mixes feeding multiple fires (the term of art for a nozzle). The typical arrangement is a cross-fire, with fires arranged in pairs pointed toward the center. In such cases its easier to use a single mixer supplying multiple fires. Its advantage is that you get automatic balancing across the pair. The disadvantage is an increased flashback issue, but it's only at most a meter of flexible hose, so this is manageable.

Now that I've outlined the basics so you might have some kind of understanding of the issue, I have to say that a 20m run of premixed gas is totally stupid and possibly crazy. Don't do it. What the actual problem you have that led to such an idea?

Panache - 17-8-2011 at 03:45

I intended to have a further needle valve at the end of the run, this would double as a flashback arrestor i guess. you have basically hit the nail on the head, i'm wanting to build a three or four head glassblowing jig and wanted to run it from my existing setup. So the actual problem is not wanting to run separate lines and this drove me to question the methodology, i assumed as much but its good to ask why sometimes, even if the answer may be obvious, thnx for replying mr fawkes, tis appreciated.