Welder Training by the Numbers
Chapter 5 of 10 · 10 min read
The cost of practice
Build the cost of an hour of welding practice from wages, instruction, plate, wire, gas, power and the booth, cost a trainee's first year, and find the savings that do not slow learning.
Practice is the core of welder training, and it costs money every hour: the trainee’s wages, the instructor’s time, the plate that becomes coupons, the wire and gas burned on them, the electricity, the grinding discs, and the booth itself. Training budgets are often set without these numbers, either generously and then cut when they are noticed, or tightly and then quietly overrun. Knowing the cost of an hour of practice lets a manager plan training honestly, compare it with the alternatives, and find savings that do not slow the trainees down.
This chapter builds the cost of an hour of practice for the fabricator’s trainees, costs a trainee’s first year, and looks at where money can be saved and where saving it costs more than it saves.
The elements of an hour
An hour of practice in a training booth consumes:
- the trainee’s time, at their wage plus on-costs;
- the instructor’s time, in proportion to how many trainees one instructor supervises;
- plate for coupons, which is cut, welded, broken and scrapped;
- wire or electrodes, burned while the arc is on;
- shielding gas, flowing while the arc is on and during purges;
- electricity, for the arc and the extraction;
- consumable parts and wear items, such as contact tips, nozzles, grinding and cutting discs, and the wear on gloves and helmets;
- the booth, its equipment, space and maintenance, spread over its hours of use.
The arc time fraction from Chapter 2, about 20% in practice, sets the wire, gas and power used. The instructor ratio from Chapter 4 sets the instruction cost.
An hour, costed
- Trainee wage with on-costs
- A$32.00
- Instruction, A$90 ÷ 4 trainees
- A$22.50
- Plate, about 3.5 kg at A$2.50/kg
- A$8.75
- Wire, about 0.8 kg at A$7.00/kg
- A$5.60
- gas, 0.18 m³ at A$9.00/m³ ≈ A$1.62; power, about 1.4 kWh at 30 cents ≈ A$0.42
- wear items ≈ A$2.50; booth equipment and space ≈ A$4.00
- total ≈ A$77.40 an hour
The wire figure comes from about 12 minutes of arc time in the hour at a wire use of about 4 kg per hour of arc time for the shop’s flux-cored wire, and the gas from 12 minutes at 15 L/min. The booth figure spreads the cost of the power source, extraction, positioners and bench over about five years of use. The plate figure assumes that each coupon is used for several runs before it is broken, and that offcuts from production are used wherever possible.
The shape of the cost is the main lesson. People’s time, the trainee’s and the instructor’s, is about 70% of an hour of practice. Plate and wire are significant but secondary, and gas and power are small. A saving on materials of 20% would cut the cost of an hour by about A$3; a saving on instruction of half would cut it by about A$11, but, as the failure at the end of this chapter shows, would slow learning by more than it saves.
Later modules cost less
As trainees become more independent, one instructor can supervise more of them. This is an illustration. In the later fillet modules, the fabricator’s instructor supervises about eight trainee hours for every hour of instruction, so instruction falls from A$22.50 to about A$11.25 an hour, and the hour costs about A$66 instead of A$77. Materials stay about the same, because the trainees burn as much wire per hour; if anything, they burn more, as their arc time fraction rises with confidence.
A trainee’s first year
The fabricator’s first-year plan (Chapter 4) has about 330 hours of booth practice, 70 hours of supervised production, 60 hours of theory sessions, and a qualification test.
| Hours | Cost | |
|---|---|---|
| Early booth practice, at about A$77/h | 150 | about A$11,610 |
| Later booth practice, at about A$66/h | 180 | about A$11,905 |
| Supervised production, at about A$53/h | 70 | about A$3,690 |
| Theory sessions, trainee and instructor share | 60 | about A$3,270 |
| Qualification test, plates, testing and witness | — | about A$1,500 |
| Total | 460 | about A$31,975 |
| Less value of supervised production work | — | about A$3,150 |
| Net first-year training cost | — | about A$28,800 |
About A$28,800 a trainee for the first year, or about A$115,000 for all four, is a significant investment for a business of twenty-two people. Chapter 10 sets it against the value of the trained welders and the alternatives. The figure is also a useful check on any training offer from outside the business: a course that claims to produce a production-ready welder in a fraction of these hours, at a fraction of this cost, deserves careful questions about what its graduates can actually do.
Costs that do not appear on the booth sheet
Some training costs never reach the training cost centre. Experienced welders who mentor trainees on supervised production work slow down while they explain, check and correct, and that lost output is a real cost. Production supervisors spend time planning work that a trainee can safely do. Jobs that use trainee welding take longer and need more inspection. This is an illustration. If each of the four trainees’ 70 hours of supervised production slows an experienced welder by about a quarter of an hour for every hour, the mentoring costs about 4 × 70 × 0.25 = 70 hours of experienced welder time, about A$3,150 at A$45 an hour. The manager adds an allowance for these costs to the training budget, so that the true cost is known, even though it cannot be measured as precisely as the plate and wire.
Cost per competent welder
Cost per hour is useful for planning, but the number that matters is the cost of each welder who reaches the target. A trainee who leaves after six months, or who never reaches the target, has cost money without producing a competent welder, and that cost is part of the price of the ones who succeed. This is an illustration. If one trainee in five, on average, leaves or fails to reach the target after spending about half the first-year cost, the cost per competent welder rises from about A$28,800 to about (4 × 28,800 + 14,400) ÷ 4 ≈ A$32,400. Choosing trainees carefully (Chapter 1), keeping them (Chapter 10) and catching a trainee whose curve has stalled early (Chapter 3) all bring that number down.
Budgeting the year
The fabricator now sets a training budget at the start of each year, from the plan in Chapter 4 and the costs in this chapter, and reviews it each quarter against the cost centre. The budget shows the owner, before training starts, what the business is committing to, and it gives the manager authority to protect the training when production pressure builds. A training budget that exists only as a general intention is the first thing cut in a busy month; one that has been costed, approved and tracked is much harder to cut quietly.
Saving on materials
There are sensible ways to reduce the materials cost without slowing learning:
- Use offcuts. Production offcuts, if they are the right steel and thickness, make good coupons at little cost.
- Use each coupon several times. A tee coupon can carry several runs on each side before it is broken, and bead practice can be done on scrap plate that would otherwise go to the recycler.
- Recycle the scrap. Broken coupons and offcuts are steel scrap with a value, not waste.
- Use smaller coupons where they teach the same thing. A 150 mm coupon teaches as much as a 300 mm one for most fillet practice.
- Keep the gas flow right. The gas flow in training booths should be set to the procedure, not higher, as Industrial Gases by the Numbers shows in its chapter on shielding and purging gases.
Simulators
Welding simulators, which use a dummy gun and a screen or headset to show a simulated weld pool and score the trainee’s distance, angle and speed, are increasingly used in the earliest stages of training. They use no plate, wire, gas or power, and they give immediate feedback on exactly the variables a beginner struggles with. Their limitation is that they do not reproduce the real pool, the heat, the noise or the feel of welding, so they shorten the cognitive stage rather than replace practice. A business considering one should cost it against the materials it would save in the first weeks and the instructor time it would free, and check with trainees and instructors whether the skills transfer to the booth.
What not to cut
Some savings cost more than they save. Instruction is the largest lever on the rate of learning, because a trainee left alone practises their mistakes. Testing is the source of the most useful feedback, and breaking fewer coupons to save plate deprives the trainee of it. Practice hours themselves cannot be cut without moving the trainee’s learning curve to the right: the hours in Chapter 3 are the hours the skill needs. And equipment that does not match production costs time later, when the trainee has to relearn. The cheapest training is the one that reaches the target in the fewest hours, not the one with the lowest cost per hour.
The cost of slower learning
A slower learning curve costs more than the practice hours alone. This is an illustration. If cutting instruction to save about A$11 an hour lengthened a trainee’s time to the target from 400 to 600 practice hours, the extra 200 hours at about A$66 would cost about A$13,200, far more than the 400 × A$11 ≈ A$4,400 saved on instruction. On top of that, the trainee would reach production welding months later, and every month of delay is a month in which the business pays for welding it cannot do, through overtime, subcontracting or late jobs.
Against the cost of hiring
The training cost is easier to judge beside the cost of the alternative. This is an illustration. When the fabricator last tried to hire a qualified welder, it paid a recruitment agency about A$8,000, offered a wage premium of about A$5 an hour above its own scale to attract the candidate, and spent about two weeks of supervision and testing confirming that the new welder’s skills matched the work. Over a year of about 1,800 working hours, the premium alone is about A$9,000, and it continues every year. A trained welder costs more in the first year, about A$32,400 on the figures above, but nothing extra after that, and was trained in the shop’s own processes and procedures. Neither option is free, and the comparison in Chapter 10 depends on how long each welder stays. The point here is that training is not an expense set against nothing: it is set against recruitment, premiums, subcontracting and the cost of work not done for want of welders.
Recording training costs
The fabricator records training costs against each trainee as a cost centre: hours in the booth from the booth log, instructor hours from timesheets, plate and wire from the training stores issue, and test costs from invoices. The record turns the illustrative figures of this chapter into the business’s own, and it lets the manager compare trainees, see whether a change in method saved money, and answer the owner’s question of what training costs with a number rather than an estimate.
A failure: the instructor who was always busy
This is an illustration. During a busy quarter, the lead instructor was needed on production almost full time, and the four trainees practised largely unsupervised, with coupons inspected and broken only at the end of each week. The training cost per hour fell by nearly a third. The trainees’ reject rates, plotted on their learning curves, stopped falling for most of the quarter; Ben’s actually rose, because a fault in his vertical technique went uncorrected for weeks and became a habit.
When the quarter ended, the manager estimated that the trainees had lost about 80 hours each of effective progress, worth about 80 × 66 × 4 ≈ A$21,000 in practice that would have to be repeated, against a saving of about A$9,000 in instruction. The plan now protects a minimum of instructor time for the trainees, even in busy periods, as Chapter 4 describes.
Exercise: costing a refresher
An experienced welder needs a two-day refresher in vertical-up welding before a site job, about 12 hours of booth practice with an instructor ratio of one to one for the first 4 hours and one to four for the rest. Taking the welder’s wage with on-costs as A$45 an hour, the instructor at A$90 an hour, and materials, wear and booth at about A$23 an hour, calculate the cost of the refresher, excluding GST.
The welder’s time is 12 × 45 = A$540. Instruction is 4 × 90 + 8 × 90 ÷ 4 = 360 + 180 = A$540. Materials, wear and booth are 12 × 23 = A$276. The total is about A$1,356.
That is less than the cost of repairing about 23 m of weld at A$60 a metre, which a welder out of practice on vertical-up welding could easily need on a single site job. Chapter 8 explains when refreshers are needed.
