Rich Smith - Welder and weld tech.

Serves "seeing is understanding".

Cornwall 2024/2025

Mainly

• mineral processing
• marine

"Channels" (beams)

Is marine.
Provided a solution they didn't known was needed but were glad they got.

overall impression of this component

Seen in the first stage of fabrication:

The main structure is a "U" channel, 5mm thickness plate "profiled" and press-brake'd.
The "notched" end-regions take a bent plate inserted, as seen.

The "channel" serves as a strut (imposes a longitudinal distance) and a beam (takes loads "normal" (at right-angles to) the length dimension).

crucial points

The crucial issue in fabricating and welding this component: the "channel" pressing would have to be something like 1-in-2000 parts accurate for the fabrication to fit-together per the drawing taken literally.
Not achievable.
A press-brake accurate enough to make the "channel" fabricating "literally as per drawing" would repay over 100's of thousands of the components. The production run is in the high hundreds. Hence the welder intervening, mainly applying spirit-levels, to get the design intent, is the most economic solution.
That time is roughly about 15 minutes per component.

Another point: this component seemingly takes far too much work for what it is. However, its form simplifies everything else about the structure's design.

pressings first fitup

Fitting the "insert plates" with the 135degree fold:

Mark-up used during the welder intervention:

Both marking-up the results of applying spirit-levels

• the "innermost" round marking containing "bars" and with the arrow, above the trestle, indicates how many packers under that location, and if there is another packer at the other trestle (if so the arrow points to where it is). If the channels get knocked the level can be remade without using the spirit-level(s) again. This intervention has distributed the "twist" error about equally between the two ends.
• the "end" markings are how much depth in millimetres to grind from the locating notch at that side. An alternative intervention is to on the opposite side pad with a run of weld metal.

Another advantage gained by these interventions is no pressings discarded to scrap.
This component cannot be seen in-service, so within structural bounds there there is no issue doing significant corrections.

pressings second fitup

Fit the bolting-surface end-plates:

A "jig"; the long square with the two plates gauging the corrent separation of the end-plates, is seen on the far left-hand-side.
After the interventions / corrections applied at the first stage, the "jig" placement and the fit of the end-plates is "literal" - the fabrication is the ideal simplest case.

A spirit level is again used in a way not shown: by an intervening correction to make the end-plates "vertical". The correction being applied with a large adjustable spanner slightly bending the tack-welds. Transverse squareness was accurately given by the "jig". No reference surface was accurate enough to act as a reference for using a square. The "jig" is checked by putting it on its side on a flat surface and using a spirit-level looking for verticals on its end-plates.

weld pressings & endplates

A lot of welding. Various conditions: "spray" and "dip" using GMAW/MIG.

Example weld - "dip-transfer" in this case.

closing plate fit & weld

These custom "wedge-clamps" were "leap-frogged" along the component. Squeeze-up the fit; weld made; clamp freed and moved ahead of the next clamp.

Example cover-plate to channel weld. Is along the length of tabs.
The weld is seen to have good fluidity and therfore fusion, as evidenced by the brown "glass" on the surface of the weld. Without excessive heat melting the edge of the "channel".

Repetitive but satisfying job.
The components unconditionally fitting during product assembly freed effort to cosmetic aspects of the product for customer appeal.

SHS endplate

Minimum space, to the millimetre, was crucial in this case.
Hence - outside-corner weld end-cap not an option (would have added another 8mm to width/length).

Not full-penetration - no engineering need - but assertive penetration, and groove enables directed powerful arc with surface tension holding fluid weld-pool.

Flushed-off after welding giving planar face along plane seen in this prep.

Pivot arms

Machine parts.
No structural redundancy. However, in-service forces / stresses are low. Also meaning dynamic stresses and fatigue resistance is a limited issue. "Simply" - well-fused welds are "it".

Weld "specification" was the prototype I offered - marked "Proto" as seen in picture.

Cornwall autumn 2022 - "outdoor" SMAW

Agency weld-fab'ing on and above deck fittings, the Foreman asked if I was confident to weld the hull. The painter found a weld pore, indeed proving full weld thickness when I ground it, and their welder was on-leave. I affirmed so.

The weld was in the "PE" position, vertically upward under the hull. Given

• defect at/due-to "Y" intersection of 3 plates
• "blowing" is certainly due to trapped moisture between plates
• welding position overhead
• at deepest displacement of the hull
• criticality of weld given this vessel does motor on the open sea

I selected "smallish" 3.2mm 7018 (calcium-carbonate based flux) rods, for the strategy

• drive quite hard at upper end of rod's Amps range
• stringer-beads stringer'ing multi-run weld
• to give assertive fusion
• with rapid freeze (positional weld & avoid blowholes)
• while focused and fluid running into the "Y" intersection
• highest strength and toughness for intersection-point
• weld structure clear in-view reassuring on this critical weld

Result:

Asked by the Foreman if the weld is good, I asked he be the judge of that. He reappeared in 3 minutes - the time to walk down and under the boat and back to the wharf - and nodded.

Prior work

Tungsten quarry 2018

Many in mineral processing in Cornwall worked on Hemerdon Hill and recognise these components. Great introduction.

GMAW / MIG.
Tilt at 45degrees puts these fillet welds in the "PA" (downward) position. Which is very insensitive to the welding torch being forced out of vertical by the space constraint between the gusset-plates.

They are "outboard" plinths for bearings of the Hemerdon mill. Hence are subject to high dynamic loadings.

Cleat 2020

Quick "EXC2" full-penetration double bevel - GMAW / MIG

MIG/GMAW aluminium 2021/2022

All marine.

I will be frank:

• I got good work using a very accurate easy-to-optimise "classic" "constant voltage" welding condition
• I have been part of unhappy experiences using "electronic" welding conditions giving extremely high rejection rates

Work done - boat-building new-build example

(R. Smith, 07Aug2025 to 08Aug2025, 11Aug2025 (7018pe))