I've been messing around trying to learn OA welding for some time now and
seem to be stuck. It's for aircraft welding so it neds to be right (thin
steel tubing) but I'm still making a lot of errors and I don't realy
understand why. I can run a fairly nice looking bead around a tube cluster,
but I end up with undercut adjacent to the weld almost every time. At the
moment I'm using some mild steel offcuts about .065 with 1/16th rod. I
believe I have the right tip and it generaly takes me about ten seconds to
get a puddle going so I think I've got the flame OK but I'm still getting
persistent undercut. If I back off the flame at all, I'm not getting a god
weld and if I increase the gas the whoe thing runs away too fast and I have
an even bigger mess.The video I have tells me to keep the flame
perpendicular to the weld and this seems to help keep the metal in front
from getting too hot, but I'm still having problems!
I can post some pics of the welds if anyone thinks that might help..
I'd love to get some real time instruction, but that's not realy an option
where I live...
I'd say to try putting in enough filler to bring the puddle up and move
forward with the heat far enough to let the rear of the puddle freeze.
Move in discrete increments with an adequate amount of filler to cool
the puddle and enough travel to take the heat off it.
Interesting. I was thining it was the exact opposite. Certainly when i use
a thicker rod the problem gets worse, probably because I'm putting too much
energy into gettng the rod melted and I'm overheating the steel. I think my
pace is OK since the weld actually looks pretty good in every other way.
mostly the bead looks good and it's nice and shiny and not pitted, but what
you say makes sense. I'll try iy now and let you know how i got on!
Tried as you suggested, but no luck. some of my welds are perfect, others
are undercut and i really can't see what i'm doing differently on each.
From the beginnning, my routine is this, I set both gasses to about 8 psi
on the regulators. I'm using a #2 tip (can't remember the name of the
torch, it's old, but I'm told it's a good one) I set the acetylene flame so
it just stops smoking and then introduce the oxy until I get the feathered
bit just to the tip of the cone. I'm pretty sure my flame is good since the
welds are nice and shiny.
I then heat the cluster paying particular attention to the tube that has no
edge and then tack the tubes together. Then I heat up the area until it's
orange around where i'm going to begin the weld (right in the "armpit" of
the tube junction) and start with a drop of rod which i work in and then
proceed to lay a bead, using a circular motion on the torch. the bead
usually runs OK, but I find that if i try and feed the rod in more quickly
than I have (1/16 rod) it sticks in the puddle. I've had some success with
weaving the rod vback and forth across the bead, but I find this gives me a
curtainy loking weld that just doesn't look that good and pretty obviously
doesn't have proper penetration. Havein said all of that, the undercut is
tiny, usually, I'd say less than 1/100th of an inch n most of them, but
it's there and can just be seen...
Maybe try without the circular motion on the torch? Try to keep the heat
narrower.. You don't really have to feed "fast", just feed enough for
the puddle to stand up a bit and then move ahead. It sounds like there's
a lot of heat outside the puddle and the puddle is hot enough to erode
at its edge.
Mmm, yes. Is it good practice to pull the torch back a bit, or does that
just spread the heat over a larger area? I think I might be a bit too rigid
n th edistance I'm keeping the flame from the puddle and am building up a
bit to much heat that way.
Have you looked at any of the aircraft welding books/videos? There is a
book and video from the EAA on welding aircraft structures (4130), the
tinman (tinmantech.com) has videos on this too. Also has a rather nifty
torch, the Meco, that is small and is more than capable of handling
If you get a chance, visit Oskosh during the EAA convention at the end
of the month. Workshops twice a day on gas welding, and the tinman has
many demos on welding (mostly geared toward aluminum), plus you can get
the aforementioned videos.
I hope you'll forgive a brief plug:
I run Technical Video Rental. We're now in our third year of renting
out instructional and craft videos online, and delivering them via
first class mail to your door.
We carry videos on welding, machining, working with sheetmetal, etc.
You can see our selection of welding videos at
If there are any questions I can answer, please don't hesitate to write
to me at
Travis J I Corcoran
I think whether pulling the torch back is going to put too much heat to
deal with into the work is going to depend a lot on the joint, some can
take more extra heat than others, but it'll still spread the heat wider
than you might need.. are you getting good work part of the way around
the joint and undercutting in a particular portion?
Again, adding filler rod should cool the puddle- not to the point of
freezing, but to where it stands up a bit and doesn't spread. If the
undercut is occuring after this moment then the puddle has gotten hot
and melted the base metal back. If it's occuring before this then you've
not added enough filler or it's just too hot in the first place.
Keep the blue cone almost touching your puddle- if you're too far back
then you'll be putting a lot of heat outside the puddle. Too small of a
flame will also put a lot of heat into your material, too big of a flame
will make for a wide puddle and, naturally, trouble with holding it and
not burning through. Given a choice, I'd take the bigger flame and move
Are you getting acceptable work at any point? If so, watch your puddle,
flame, speed where it's working good and compare to where it isn't..
Rather than worrying about how many psi to use, why not just fiddle with
the setup until the torch burns right. I am an absolute beginner, but I
have yet to find a torch could not be set properly without having any
idea how many psi the manufacturer called for. Here's how:
1) Open the valves on both tanks, making sure both regulators are closed.
2) Open the torch valves all the way. At the start, you will be
adjusting the flame directly from the regulators.
3) Crack the acetylene regulator, and light the torch.
4) Add acetylene pressure until the flame just stays on the tip.
5) Add a tiny bit of oxygen. This will undoubtedly blow out the flame,
so back off on the acetylene until you can light it and have the flame
remain on the tip.
6) Now keep adding oxygen and backing off the acetylene until you have
the largest possible neutral flame (watch the inner cone) that will stay
on the tip. Those are the regulator settings you want for that tip.
Write them down.
7) From here on out, you can adjust the flame normally, by using the
valves on the torch.
It works for me.
One other hint: Some torch manufacturers space their tip sizes
inconveniently for working on airplanes. One tip will be too small for
welding 0.035-wall tubing, while the next size up is too big. Get an
extra small tip, and open it out with a #72 drill. The result should be
just about perfect.
Being an admitted "absolute beginner" you might want to seriously consider
taking a basic course in oxy-acetylene welding or at the very least read a
tutorial online, such as
procedure for lighting and adjusting your torch is so contrary to 100+ years
Been there, done that, found that 1psi acetylene doesn't always work. In
fact, in my limited experience, it seldom works. The pressures required
for each tip vary a whole lot more than that chapter would have you
believe, just as the tip sizes vary widely from one manufacturer to
another, even though the number may be the same. For example, the
Henrob torch takes 4 psi for both oxygen and acetylene, no matter which
tip you use (with the possible exception of the cutting torch; I've
never used it.) My Harris set requires different settings for each tip,
with different pressure for oxygen and acetylene on a given tip--this
even though it has special tips designed to produce a Henrob-style
I came up with this procedure myself some years ago, when trying to get
used to the Henrob torch after having used mostly a Victor and working
with regulators I didn't trust. However, in looking for the pressure
requirements for various tips a few moments ago, I found the following,
at http://www.tinmantech.com/html/faq_welding_tip_pressure.html :
"Smith man asks: 'I have one question about the Smith outfit. Most of
the other outfits (non Smith) I have used say to use a Acetylene and
oxygen pressure of 2 to 6 psi depending on the size material and size of
the tip. The Smith instructions tell you to set the Acetylene and oxygen
at 10psi for the welding tips from 32ga up to 1/2" (mw200 series) and 10
psi Acetylene for all of the sizes of the cutting tips. (mc12 series)
the oxygen pressures varied per size cutting tip as it should. Why are
the pressures so high for the welding tip? Mainly the Acetylene?? (kinda
close to 15) I have never used a setting that high before! I called
Smith and they told me "yep ...10 psi Oxygen and Acetylene!"
"Tin Man answers:
"Hmmmmm, I would try opening the bottles with the regs closed. Open the
torch valves all the way, and then slightly open the acet reg. and the
ox.reg and light the torch. Then using the regs only, set for the best
flame.Now check the regs for the pressure... 2-3 lbs for a .035 tip?
Sounds normal to me."
Seems like Smith disagrees with 1 psi acetylene, too, but it was the Tin
Man's answer that interested me. If my advice wasn't any good, it looks
like I am at least in good company.
Try it; you'll like it.
I think the most valuble thing that I took away from welding school was
taught in the very first few hours during our safety orientation. We were
shown a film dealing with what can go wrong when using or misusing
oxy-acetylene equipment. After seeing that film, this old boy has been
convinced that making up your own procedures is nothing but foolhardy.
Passing such "knowledge" on to others is, IMHO nothing short of
The correct way to light and adjust a torch is by following the
manufacturer's directions. Period.
Well, I'll also light (I mean line!) up behind Owen. A proper flame is
a proper flame and there's no hiding it.
As to 100 years of standard procedure, I hate to use a politically
charged metaphor, but slavery was once considered "standard procedure".
Be prudent but don't be afraid to try something new. At least that's
my philosophy. I have copied Owen's instructions for future reference.
Way to go Owen!
P.S. If I blow myself up I hereby indemnify and hold you harmless.
Gentlemen, start your torches!
Thanks, Vernon. And I'll stand by what I learned the other day: If Kent
White recommends it, it can't be all bad, even if this bungler did
figure out independently. This side of Ernie, I can't think of anyone
I'd trust much more when it comes to welding, and O/A is about all the
Tin Man uses.
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.