This post to sci.engr.joining.welding newsgroup covers the topic of setting your gas pressures and selecting the right nozzle size.
These things which a lot of people starting on Oxy-acet. get troubled about is what pressure to set on the regulators for the separate supplies of oxygen and acetylene and what nozzle size to fit on the torch for a given job. I'll quote verbatim a "followup" post I made on s.e.j.w. covering both these points.
Thread title was "Annoying welding problem":
My own addition (experience - UK City & Guilds training, plenty of use
oxy-acet at work)
Nozzle size. You want a certain amount of heat, so you will try to put
a certain mass per unit time of oxygen and acetylene through the nozzle,
regardless of its size, so
- too small and the gas velocity is too high and it blows the molten
metal away - as soon as the metal melts it sprays away on the far side,
so you end up reducing the size of the object you are heating but never
welding it, at the extreme
- too large and the gas velocity is lower than the flame velocity and
you get light-backs into the inside of the nozzle, with noted bangs and
the weld pool being thrown around, usually over you, especially in your
hair.
The nozzle size if fairly critical. By that I mean, if the job rightly
calls for a #3, you would be happy neither with a #2 nor a #4.
Another reason for "banging" is that the nozzle simply gets so hot that
there is lighting of the mixed oxygen and acetylene in the inside of the
nozzle. If you really do have to work for a long time in an awkward
corner, it is OK to dip the nozzle of the torch in a bucket of water
every now and again. Told flame will stay lit while you do this, but
never tried it...!
Pressure at the regulator.
I don't think it will have any effect on the lighting-back. The reason
is the given-massflow argument regarding the amount of heat you need at
the weld. If you raise the pressure out of the regulator, you would
restrict back the flow in proportion with the valve on the torch. So
the pressure in the nozzle will be exactly the same! (surely?). I tend
to set pressure so that the torch valve is quite well open and a lot of
movement makes a small adjustment. That gives you easy fine adjustment.
For goodness sake, you can set the acetylene pressure at the regulator
by setting the torch adjustment valve wide open and having the regulator
set way low, light the acetylene and turn up the acetylene regulator
pressure until the flame is a bit bigger then the flame you know you
will need to weld. Then you restrict back on the torch valve to the
correct flame. Same with oxygen. You will likely end up with pressures
less that 8 and 8 (psi, that is). That will give you fine control in
the range you want. This works because, burning acetylene only, you
identify the range the nozzle will be correctly drive at because it is
when, increasing the acetylene supply, the acetylene-only flame stops
being smokey and becomes a clean roaring incandescent yellow/white. I
have to admit the main reason I do this is to be a flash bastard and
show off by operating the torch very rapidly wiht one hand, using only
the fingers of the hand it is held in. You can "drive the nozzle hard"
by increasing the acetylene supply beyond the point where the
acetylene-only flame burns clean. With the oxygen on and a proper
"neutral" flame, the gas velocity will be higher and you can actually
use this. I found it helpful in some fillet joints, where you might
drive a #2 hard rather than go to a #3 driven at normal rate, for the
same heat.
Haven't used oxy-acet much for a while. Would have thought try #3. #2
would be too small for 1/8th inch, for sure (?).
BTW - filling technique - you know that "melting off the filler rod like
sealing wax for a letter", with metal dripping into the melt pool, is
definitely a defect in oxy-acet? You get an oversized uncontrolled
overheated looking weld if you do this, compared to the controlled
fine-looking weld with desired profile if you dip ("harpoon") the rod
into the weld pool as needed far slightly afar.
I will gladly be stood corrected if any of the above is not correct.
Richard Smith
No-one corrected or qualified this post of May 2004, so likely information regarded as good.