Success hangs on having the right rod and enough current. Fast-freezing slag is called for, which will stay above the arc rather than running down under it and getting in the weld pool. ESAB’s OK 46.00 is a good choice, and is equally at home welding on the flat or vertically upwards.
Vertical down welds need more current than the same weld on the flat, as the extra force in the arc helps keep the molten pool clear of slag. Start with a setting close to the top of the rod’s recommended range. Center the rod over the joint line (1.48) and keep it just below horizontal. Travel fast enough to stay ahead of the descending slag curtain. Don’t let the arc get too long or weld metal won’t find its way to the pool.
Welding vertically up produces deep penetration and apprehension in equal measures. But with the plant set right and the technique sorted, worry will go away.
Electrode choice isn’t as critical as it is with vertical down. Most general-purpose mild steel rods will work happily upwards, as gravity carries off the slag.
Heat control is everything. Keep the current down, dialling in 10-15% less than on the flat. But don’t go too low or the rod will stick and slag traps form in cool areas. It’s a balancing act: experiment at the lower end of the rod’s recommended range, settling on the lowest current compatible with trap-free, easy-running work.
While welding you’ll see the arc digging a crater and molten metal draining away under gravity. This is where vertical up’s fine penetration comes from. But if the area overheats, the end result is a series of dangling blobs like melted candle wax (1.52).
The first pass in a multi-run joint is made without weaving, unless you need to control burn-though. Subsequent runs rely on a weave to support the molten metal (1.49). Each lateral weave creates a horizontal ledge, which is used as a step to hold fresh weld. It’s like making each rung of a ladder as you climb.
1.49. A two-pass vertical up joint showing different weave patterns. The first run used a triangular weave, pushing the tip into the root gap until a change in arc noise indicated full penetration. The ‘keyhole’ shape in the penetration gap confirms this (arrow). Pausing at the borders of the weave (circles) filled in the edges. The second pass features an easier H-pattern weave, again with edge pauses. But the second weave wasn’t wide enough to blend with the plate’s outer edges, and there is some undercut near the start caused by too-brief pauses.
1.50. A simple crescent weave features in this quality single-pass run. Neatness and strength come from very deliberate rod tip movement, coupled with a pause at each edge to let the crater fill (circles).
1.51. Assorted weave patterns. Use number 1 for flat (down hand) work, 2-5 for vertical up. Choose whichever best suits the job in hand.
Follow the arrows in pics 1.49, 1.50 and 1.51 to see the track the rod tip takes during different weaves. A good upward weld has a wrinkled, prune-like finish; it’s the result of all those individual ladder rungs. One or more downward passes using a crescent weave can tidy things up if needs be, but reset the current and use the right rod (page 31).
For best weave control imagine you’re a slow-motion robot, constantly tracing the same pattern on the joint at the same speed. Keep the rod horizontal; or in thick sections, just above horizontal. Center it over the joint to balance fusion between the plates, and hold arc length as short as current allows to direct metal where you want it.
If the set has more than one output voltage, use the higher to help arc establishment and smoothness when welding vertically up. Relax, taking the electrode cable’s weight on surrounding metalwork and propping yourself comfortably. When you’re welding, try keeping a light grip on the electrode holder. Enjoy yourself! Repeat after me: vertical up is fun, vertical up is fun….
1.52. Too much heat turns a vertical up weld into a mess. Gravity pulls molten metal down too far before it solidifies (arrow).
Overhead welds are another tightrope act. It’s flat welding turned upside down, only this time gravity is working against you. Metal will transfer across the arc thanks to magnetic forces, but the molten pool wants to drip. So three rules apply: Keep the arc as short as you can to lessen the chance of losing metal during transfer; keep the current as low as possible so weld metal can freeze before it sags; and travel as fast as you can so heat build-up is low. All this has to be balanced against achieving good fusion and penetration – hence the tightrope act. A visual clue on how well you’re doing is in the bead shape. If the center hangs down, you’re a little too hot or too slow.
SMAW HELPLINE
Everybody (but everybody) makes mistakes. There’s no shame in it. But if you want to improve it helps to be able to spot faults, assess their seriousness and know where you went wrong. Pictures A-H deal with common problems and ways around them. (All pictures courtesy of The Welding Institute).
A. Slag traps (arrow) are a pain. Avoid them by keeping the four variables right — use enough current and the appropriate rod angle, hold arc length fairly short and don’t travel too fast. The one above came from using too little heat or a long arc. Rod metal took the shortest path to the vertical plate, leaving a hole for slag to fill. Dealing with traps is covered on page 30.
B. This weld seemed fine on the surface, but a peek inside shows unequal length legs and a slag trap instead of root fusion. Low current, a too-long arc and the wrong rod angle were the cause.
C. This T-fillet, welded with the right current, rod angle and arc length, shows equal leg lengths and good fusion in both plates. A second, weaved run was added – spot its coarse, grainy look. The first run has been normalized by the heat of the second, leaving its grains smaller and improving strength. The small hole at the root isn’t a problem: it comes from the vertical plate’s original rounded profile. If you’re not convinced, lay a rule along the right-hand edge.
D. Two faults in one here: incomplete penetration despite the gap between plates, plus a circular slag inclusion at eight o’clock. A smaller diameter rod, run at a reasonable current, would have helped penetration and provided enough heat to prevent trap formation. Just winding up the current on a big rod wouldn’t do, as its girth would still stop it reaching the bottom of the joint.