What are the main categories of yield loss in a bar rolling mill?

Yield losses in a bar mill fall into six main categories: crop losses (40–50% of total losses) from head and tail sections removed from each billet; cobbles and mill stoppages (15–25%); dimensional and quality rejects (10–20%); scale and oxidation losses (8–12%) from high-temperature reheating; scarfing and conditioning of billet defects (5–10%); and sampling for quality testing (2–4%). Each category requires different improvement strategies.

What is the financial impact of a 2% yield improvement in a bar mill?

For a 500,000 tonne per year bar mill purchasing billets at €500 per tonne and selling bars at €650 per tonne, a 2% yield improvement from 92% to 94% recovers 10,000 tonnes of saleable product. After accounting for lost scrap revenue, this improves the bottom line by approximately €4 million annually. At higher bar prices of €800 per tonne, the same improvement generates approximately €7 million. The key insight is that the marginal cost of the recovered tonnage is nearly zero — it flows almost directly to profit.

How does batch size affect rolling mill yield?

Larger batch sizes significantly improve yield because crop losses are a fixed overhead per billet — the head and tail crops represent a smaller percentage of total tonnage when batches are larger. Additionally, larger batches stabilise furnace temperatures (reducing required crop lengths), improve threading consistency (reducing cobbles), and allow operators to establish better rhythm. Mills running batches of 10–20 tonnes typically achieve 89–92% yield; mills running 100–200 tonne batches achieve 93–96% yield. Batch size optimisation requires no capital investment, only production planning discipline.

What is furnace atmosphere control and why does it reduce scale losses?

Furnace atmosphere control manages the air-fuel ratio in a reheating furnace to minimise free oxygen in contact with hot steel. In a conventional furnace with uncontrolled combustion, excess air is introduced to ensure complete combustion — but this excess oxygen causes heavy scale (iron oxide) formation, typically 1.5–2.5% of billet weight. By running at stoichiometric or slightly sub-stoichiometric combustion, using zone-specific oxygen sensors and control algorithms, scale losses can be reduced to around 1.0%. On a 500,000 tonne bar mill, this saves 7,500 tonnes per year. Implementation costs €200,000–€400,000, with payback of 6–12 weeks.

What is a typical improvement roadmap for steel mill yield optimisation?

A systematic yield improvement roadmap has four stages. First, measure current state by loss category — weekly tracking broken out by crops, cobbles, rejects, and scale losses, benchmarked against best-practice mills. Second, capture quick wins in 3–6 months with minimal capital: batch size optimisation, crop length reduction through better temperature tracking, and operator training. Typical gain: 0.5–1.0 percentage points. Third, implement medium-term capital projects in 6–12 months: furnace atmosphere control, improved descaling, and mill automation upgrades. Typical gain: 1.0–2.0 percentage points. Fourth, strategic investments over 1–2 years: comprehensive furnace rebuilds and advanced process control systems. Typical gain: 2.0–3.0 percentage points.

Steel Mill Yield Losses: Hidden Costs & Profit Recovery

STEELONTHENET PODCAST

Episode 006: Understanding Steel Mill Yield Losses

From Hidden Costs to Profit Recovery

2 April 2026 19 minutes Dr Andrzej M Kotas

Listen to Episode

0:00 / 0:00

Download: MP3 file | Duration: ~19 min | Author credentials: ORCID iD

Episode Overview

Yield losses in a bar rolling mill are the invisible drain on profitability that never triggers an alarm. Unlike equipment breakdowns or quality complaints, they are embedded in daily operations — hidden in crop bins and scrap buckets, accepted as normal. But they are not inevitable. They are identifiable, measurable, and improvable.

This episode examines exactly where yield losses occur in a typical bar mill, quantifies their financial impact, and identifies the improvement opportunities that can transform your bottom line. We look at crop losses, cobbles, quality rejects, and scale formation. We explore how batch sizing affects yields. And we examine furnace atmosphere control — a technology that can cut scale losses by more than half, with payback periods measured in weeks rather than years.

For a 500,000 tonne per year bar mill, a two-percentage-point yield improvement generates €4–7 million in additional profit annually. That is not revenue. That is pure profit flowing straight to the bottom line.

Key Takeaways

Yield is often the single most important operational metric after safety: A 2% yield improvement on a 500,000 tonne bar mill generates €4–7 million in additional profit depending on product pricing. Some improvements require zero capital investment.
Yield losses fall into identifiable categories: Crop losses (40–50% of total), cobbles and mill stoppages (15–25%), dimensional and quality rejects (10–20%), scale and oxidation losses (8–12%), scarfing and conditioning (5–10%), and sampling (2–4%). Each requires different improvement strategies.
Batch size optimisation is zero-capital low-hanging fruit: Moving from 20-tonne batches to 150-tonne batches can improve yield by 1–2 percentage points through more stable furnace temperatures, improved threading consistency, and better operator rhythm.
Furnace atmosphere control delivers exceptional returns: Cutting scale losses from 2.5% to 1.0% on a 500,000 tonne mill saves 7,500 tonnes — €3.75 million annually. Implementation cost of €200,000–€400,000 gives payback of 6–12 weeks.
Approach yield improvement systematically: Measure current state by loss category, capture quick wins first (0.5–1.0 point gain in 3–6 months), then progress to medium-term capital projects and strategic investments.
External expertise accelerates improvement: Familiarity breeds blind spots. A specialist who has benchmarked dozens of mills can identify opportunities that internal teams consistently overlook.

Full Transcript [Show]

Introduction (00:00–02:00)

Welcome to SteelOnTheNet Podcasts. I'm Dr Andrzej M Kotas, and today is April 2nd, 2026.

Today we're examining something that quietly drains millions from steel mill profitability every single year: yield losses in bar rolling mills.

Let me start with a scenario. You're operating a 500,000 tonne per year bar mill. Your yield is running at 92%. Management thinks this is reasonable performance.

But here's what they don't realise: if you could improve that yield to 94% — just two percentage points — you'd generate an additional €5 million in profit annually. That's not revenue. That's pure profit flowing straight to your bottom line.

The problem is that yield losses are invisible. Unlike equipment breakdowns or quality complaints, they don't trigger alarms. They're embedded in your daily operations, hidden in crop bins and scrap buckets, disguised as "normal losses." Management accepts them because they've always been there.

But yield losses aren't inevitable. They're identifiable, measurable, and most importantly — improvable.

Over the next nineteen minutes, we're going to break down exactly where these losses occur in a typical bar mill, quantify their financial impact, and identify the improvement opportunities that can transform your bottom line. We'll examine crop losses, cobbles, quality rejects, and scale formation. We'll look at how batch sizing affects yields. And we'll dive into furnace atmosphere control — a technology that can cut your scale losses by more than half.

Let's start with why yield matters so much to your business.

Segment 1: The Financial Stakes (02:00–05:00)

Here's the brutal truth about yield losses: they hit you twice.

First, you lose the revenue. Every tonne that goes to scrap instead of becoming saleable product is revenue you'll never collect. But second — and this is what many operators miss — you've already paid for the raw material, the energy, the labour, and the overhead to process that tonne. All those costs are sunk. You get nothing back.

Let me quantify this with real numbers. Take that 500,000 tonne per year bar mill. Let's assume you're purchasing billets at €500 per tonne. Your finished bars sell for €650 per tonne.

At 92% yield, you're losing 40,000 tonnes per year to scrap. That scrap is worth perhaps €250 per tonne — half the value of your billet cost. So you've lost €250 per tonne on 40,000 tonnes. That's €10 million in lost value annually.

Now improve your yield to 94%. Your losses drop to 30,000 tonnes. You've recovered 10,000 tonnes that can now be sold as finished product. That's €6.5 million in additional revenue. Subtract the €2.5 million scrap value you would have received, and you've improved your bottom line by €4 million.

But here's what makes yield so powerful: the marginal cost of that recovered tonnage is nearly zero. You're not buying more billets. You're not running the mill longer. You're simply losing less of what you already paid for. That €4 million flows almost directly to profit.

And this calculation assumes modest steel prices. In a strong market where bars sell for €800 per tonne, that same 2% yield improvement generates €7 million in additional profit. In steel, where net margins often run 5 to 8%, yield improvement can be the difference between a mediocre year and an exceptional one.

Segment 2: Anatomy of Yield Losses (05:00–09:00)

Yield losses in a bar mill fall into several distinct categories, and understanding this breakdown is essential because each category requires different improvement strategies.

Crop Losses — The Biggest Category. Crop losses typically account for 40 to 50% of your total yield losses. These are the head and tail sections you remove from each billet or rolled bar.

Why do we crop? The head of the billet cools quickly in the furnace, developing surface defects and temperature non-uniformity. You can't thread a cold, defective head through your mill guides reliably. So you crop it — typically 150 to 300 millimetres on the head end. The tail has similar issues plus potential mechanical damage from previous processing. You'll crop 100 to 200 millimetres from the tail.

Here's the critical insight: crop losses are a percentage of your throughput. Roll in small batches, and you're making a disproportionate number of crops. Roll large batches in sequence, and the percentage drops dramatically. We'll return to this point.

Cobbles and Mill Stoppages account for 15 to 25% of total losses. A cobble — when the bar tangles or jams in the mill — often forces you to scrap not just the immediate piece but several bars upstream and downstream. There's frequently a cascade effect: one cobble disrupts your thermal rhythm, which increases the probability of the next.

Dimensional and Quality Rejects represent 10 to 20% of losses. Bars outside diameter tolerance, excessive ovality, straightness problems, surface defects — all become scrap or downgraded material. Root causes include mill setup issues, equipment wear, and material variation.

Scale and Oxidation Losses account for 8 to 12% of total losses. When you reheat steel to 1,200°C, it oxidises. Typical scale loss in an uncontrolled furnace atmosphere runs 1.5 to 2.5% of billet weight. On a 500,000 tonne operation, 2% scale loss is 10,000 tonnes per year — €5 million at €500 per tonne billet cost.

Other categories include scarfing and conditioning of billet defects (5–10%) and sampling for quality testing (2–4%).

A well-run modern mill might achieve 93 to 96% yield. Average operations run 90 to 93%. Mills with challenges struggle along at 85 to 90%.

Segment 3: The Batch Size Effect (09:00–12:00)

Here's a lever that requires zero capital investment: batch size optimisation.

The percentage of production going to crops is inversely related to batch size. Let me walk through the mathematics. Assume you're rolling 20mm diameter rebar from 130mm square billets, 12 metres long. Each billet weighs roughly 1.5 tonnes. You crop 200mm from the head, 150mm from the tail.

Roll in batches of 20 billets — 30 tonnes total — and you make 40 crops. At an average 175mm per crop, you're scrapping roughly 600 kilograms, around 2% of the batch going to crops alone.

Now roll 100 billets in sequence — 150 tonnes total. You make approximately 102 crops. But with larger batches, furnace temperatures stabilise, reducing the crop length you actually need. Mill threading becomes more consistent, reducing threading failures. Operator rhythm improves, reducing cobbles. Your actual crop lengths decrease, and your quality improves.

In practice, mills running small batches of 10 to 20 tonnes typically achieve 89 to 92% yield. Mills running batches of 100 to 200 tonnes achieve 93 to 96% yield. The sweet spot is usually 100 to 200 tonnes of the same section and grade.

Batch size optimisation is low-hanging fruit. No capital required, just production planning discipline.

Segment 4: Furnace Atmosphere Control (12:00–15:30)

Earlier I mentioned scale losses of 1.5 to 2.5% in typical reheating furnaces. Let me explain why this happens and how atmosphere control cuts these losses dramatically.

Steel oxidises at high temperature. At 1,200°C in your reheating furnace, iron reacts with oxygen to form iron oxides — scale. In a conventional furnace with uncontrolled combustion, you're introducing excess air to ensure complete fuel combustion. That excess air means excess oxygen in contact with hot steel. Result: heavy scale formation. You're literally burning away 2 to 3% of the steel you paid for.

The solution is to control your air-fuel ratio to minimise free oxygen in the furnace atmosphere. Instead of excess air, you run at stoichiometric or slightly sub-stoichiometric combustion, creating a reducing or neutral atmosphere. Modern furnaces use zone-specific control with oxygen sensors in each zone, continuous air-fuel ratio adjustment, and often inert gas injection to dilute oxygen concentration.

Let's quantify the economics for our 500,000 tonne per year bar mill. Without atmosphere control: 2.5% scale loss equals 12,500 tonnes per year, at €500 per tonne billet cost — €6.25 million lost value annually. With atmosphere control: 1.0% scale loss equals 5,000 tonnes — €2.5 million lost value. Your savings: 7,500 tonnes recovered, worth €3.75 million per year.

Implementation cost for a multi-zone walking beam or pusher furnace? €200,000 to €400,000 for the atmosphere control system. Payback at €3.75 million annual savings: 6 to 12 weeks. That's not a typo.

This is why atmosphere control is often the single highest-return investment a bar mill can make.

Segment 5: The Improvement Roadmap (15:30–17:00)

Given everything we've discussed — crop optimisation, batch sizing, atmosphere control, cobble reduction — how do you actually implement improvements? Here is a practical framework.

Step One: Measure Your Current State. You cannot improve what you don't measure. Start with comprehensive yield tracking by loss category. Break out crops, cobbles, quality rejects, scale losses, and other categories. Track weekly, not monthly. Benchmark yourself — if you're running at 90% and best-practice mills achieve 95%, that 5-point gap represents your opportunity: 25,000 tonnes and potentially €10 million in recovered value on a 500,000 tonne operation.

Step Two: Quick Wins (3–6 months, minimal capital): batch size optimisation, crop length optimisation through better temperature tracking, and operator training on threading consistency. Typical yield gain: 0.5 to 1.0 percentage points.

Step Three: Medium-Term Projects (6–12 months): furnace atmosphere control, improved descaling systems, and mill automation upgrades. Typical yield gain: 1.0 to 2.0 percentage points.

Step Four: Strategic Investments (1–2 years): comprehensive furnace rebuilds with state-of-the-art combustion control, advanced process control systems, and equipment replacement where wear is driving quality losses. Typical yield gain: 2.0 to 3.0 percentage points.

A one or two-day plant visit by someone experienced in yield optimisation can identify the dominant loss categories, quantify improvement potential, and prioritise actions by return on investment. The SteelOnTheNet team conducts exactly these assessments, bringing comparative data from mills across multiple continents. Contact us if you'd like to discuss your specific situation.

Conclusions (17:00–19:00)

Let me leave you with the critical insights about mill yield losses.

Yield is often the single most important operational metric after safety. A 2% yield improvement on a 500,000 tonne mill generates €4 to €7 million in additional profit. That's more impact than most capital projects deliver, and some improvements require zero capital.

Yield losses aren't mysterious. They break down into identifiable categories, each with specific improvement strategies. Measure which categories dominate your losses, then prioritise accordingly.

Batch size optimisation is low-hanging fruit: moving from 20-tonne batches to 150-tonne batches can improve yield by 1 to 2 percentage points with no capital investment.

Furnace atmosphere control delivers exceptional returns. Cutting scale losses from 2.5% to 1.0% generates millions in annual savings with payback periods measured in weeks.

Don't approach yield improvement piecemeal. Commission a comprehensive assessment, build a prioritised roadmap, and attack quick wins immediately while planning medium-term projects.

The mills that systematically improve yield gain competitive advantage. They generate higher margins from the same assets. They outperform competitors even in tough markets. Start measuring. Benchmark yourself. Then execute systematically.

This has been a SteelOnTheNet Podcast. I'm Dr Andrzej M Kotas. For full show notes and related resources, visit steelonthenet.com/insights/podcasts. Thank you for listening.