Surface Defects in Stamping: Scratches, Orange Peel and Roll Marks
Short answer: Surface defects in stamping come from three different places, and mixing them up wastes weeks. Scratches are usually mechanical — die clearance, burrs, scrap dragging, or handling. Orange peel is a material and forming issue: coarse grain, wrong temper, or too much stretch over a tight radius. Roll marks are inherited from the coil itself or transferred by the feed rolls and die entry. In practice, roughly 70% of cosmetic rejects we see at BQUQ trace back to die maintenance and material condition rather than the press. Fix the source, not the symptom: verify coil temper and grain, polish and re-coat die radii, control clearance to 5–8% of thickness per side, and define a cosmetic acceptance limit before the first article is signed off.
Why Do Surface Defects Matter More Than Dimensional Ones?
A stamped bracket that is 0.03 mm out of position often still assembles. A stamped terminal with a visible scratch on a gold-plated contact surface may be scrapped outright — by your customer, after the part is already in a housing.
That asymmetry is what makes cosmetic control in stamping so expensive when it is handled late. Dimensional tolerances are written on the drawing. Cosmetic limits usually are not, so they get negotiated after the first shipment, when the tool is already cut and the process is already running.
There are three practical consequences:
- Yield loss is invisible in the quote. A die that produces 3% cosmetic rejects on a 500,000-piece order is 15,000 scrapped parts, plus the inspection labour to find them.
- Defects cluster. Scratches rarely appear at 1% evenly across a run. They spike after a die crash, after a coil change, or after a new operator loads the strip.
- Plating amplifies everything. A 2 µm scratch on bare brass becomes a visible plating skip or a corrosion initiation site once nickel or gold is applied. See our notes on plating thickness control for how surface condition and coating interact.
The fix is to name the defect correctly first. "It looks bad" cannot be engineered against.
Scratches: Mechanical Damage With a Traceable Source
Scratches are linear material displacement — material is moved, not just marked. Under magnification, a scratch has a raised lip on at least one side. That distinguishes it from a roll mark, which is a smooth depression with no displaced material.
The five common sources
1. Die clearance too tight or too loose. Tight clearance shears with secondary cracking and throws fine debris; loose clearance pulls material and creates a rough, torn edge that then drags.
2. Burrs on the die block or stripper plate. A 0.02 mm raised burr on a stripper plate will scribe a line down every part that passes under it.
3. Scrap and slugs. In a progressive die, a slug that fails to clear the die window rides the strip and gouges the next station. This is the single most common cause of sudden scratch spikes.
4. Feed rolls and guides. Worn or contaminated feed rolls mark the strip before it ever reaches the die. Check the strip entering the press, not just the finished part.
5. Handling and packaging. Parts tumbling into a bin scratch each other. Plated parts are far more sensitive.
Diagnosing direction
The direction of the scratch tells you where to look. Scratches running along the feed direction are usually die or feed-roll related. Scratches running across the strip, or at an angle, usually come from handling, ejection, or a part that rotated in a bin.
| Scratch direction | Most likely source | First check |
|---|---|---|
| Parallel to feed | Feed rolls, strip guides, die entry, stripper burrs | Strip surface before station 1 |
| Parallel to feed, one station only | Local die burr or trapped slug | Die window clearance, slug evacuation |
| Random / angled | Ejection, part drop, bin handling | Chute, conveyor, packaging method |
| Concentric or curved | Draw radius galling | Radius polish, lubricant film |
| On one side of strip only | Guide misalignment, coil camber | Guide parallelism, coil edge condition |
Fixes that hold
- Deburr and polish all strip-contact surfaces to a defined Ra, then re-coat (TiN, TiCN, or DLC) where abrasive wear is expected.
- Set clearance per material: typically 5–8% of sheet thickness per side for mild steel, higher for stainless and aluminium.
- Add or restore slug clearance and, where needed, a slug-retention feature so scrap cannot ride the strip.
- Introduce a strip wiper or felt pad before the die entry to remove coil dirt and metal fines.
- Change packaging: layer separators, stack flat, no bulk dumping for cosmetic parts.
Orange Peel: A Material and Forming Problem
Orange peel is a rough, dimpled surface that looks like citrus skin. It appears when the surface grains of the material deform unevenly — some grains yield before others, and the surface roughens instead of stretching smoothly.
It is almost never a die geometry fault on its own. It is a material condition problem triggered by forming strain.
Root causes
- Coarse grain size. ASTM grain size 7 or finer is typical for good cosmetic forming. Coarse grains (5 or below) roughen early.
- Wrong temper. Full-hard or half-hard tempers have low remaining elongation. Forming them over a tight radius produces orange peel and often cracking.
- Excessive stretch over a tight radius. If the inner bend radius is below roughly 1× material thickness for a cosmetic surface, expect roughening.
- Insufficient or wrong lubricant. High friction concentrates strain at the radius.
- Anisotropy. Strong rolling texture makes the surface roughen differently in different directions.
What to do
| Action | Effect | Trade-off |
|---|---|---|
| Specify finer grain (ASTM 7–9) | Directly reduces roughening | Slight cost premium, availability lead time |
| Move to a softer temper (e.g. H1 → O or 1/4 hard) | Restores elongation | Lower yield strength, may need design change |
| Increase inner bend radius | Reduces local strain | May conflict with the assembly envelope |
| Improve lubricant / add drawing compound | Lowers friction, spreads strain | Cleaning step required before plating |
| Anneal between forming steps | Recovers ductility | Extra process, cost, possible distortion |
| Pre-polish the blank | Masks shallow roughening | Does not fix the cause; adds cost |
For a cosmetic cover or visible bracket, the cleanest fix is usually specifying the material correctly at the quotation stage. Retrofitting a softer temper after the die is cut can change springback and force a die rework — a costly sequence. Our guide to sourcing custom stamping in China covers how to write material and finish requirements so this is settled before tooling starts.
Roll Marks: Inherited From the Coil, Transferred by the Tool
Roll marks are smooth, repeating depressions or bands. They come from two directions: the mill that rolled the coil, and the rolls inside your press.
Coil-origin roll marks
These are present on the incoming strip and are simply carried through the die. Typical forms:
- Longitudinal bands from work-roll wear or chatter at the mill.
- Periodic transverse marks from roll eccentricity — spacing equals the roll circumference.
- Edge marks from coil handling and mandrel contact.
You cannot remove these in the die. You can only reject the coil, specify a surface-critical coil grade, or design the part so the marked area is not cosmetic.
Press-origin roll marks
- Feed roll marks. Worn, dirty, or over-pressure feed rolls imprint the strip. Spacing matches the feed length or roll circumference.
- Die entry and guide marks. A tight or misaligned guide block burnishes a band along the strip edge.
- Stripper plate marks. Insufficient stripper pressure or a warped plate leaves a repeating pattern at each stroke.
- Draw bead marks. Deliberate draw beads leave controlled marks — fine if the surface is hidden, unacceptable if it is visible.
Distinguishing coil marks from press marks
Run a short trial with a strip section that is known-good and visually inspected before entry. If the mark appears after the die, it is press-origin. If it is already on the incoming strip, it is coil-origin. This takes twenty minutes and saves days of argument.
| Characteristic | Coil-origin | Press-origin |
|---|---|---|
| Present on incoming strip | Yes | No |
| Spacing | Roll circumference, or continuous | Feed length or stroke-related |
| Changes with coil lot | Yes | No |
| Changes with feed settings | No | Yes |
| Removable by die polish | No | Often yes |
| Corrective action | Incoming inspection, supplier spec | Die and feeder maintenance |
A Practical Cosmetic Acceptance Standard
Most cosmetic disputes come from an undefined limit. "No visible scratches" is not a specification. Define it with three parameters:
1. Viewing condition — distance, lighting (e.g. 500 lux diffuse), viewing angle, and whether bare or gloved hands are used.
2. Limit sample — a physical boundary sample, signed by both parties, showing the worst acceptable part.
3. Quantified criteria — maximum scratch length, width, depth, count per part, and allowed zones (functional vs cosmetic).
A workable default for many industrial stamped parts:
- No scratches deeper than 10% of material thickness.
- No scratch longer than 5 mm on visible surfaces.
- Maximum 2 scratches per visible face, none crossing a sealing or contact surface.
- No orange peel visible at 300 mm under 500 lux.
- No roll mark exceeding 5 µm depth on visible faces.
These are indicative starting points. Adjust for your application — a connector contact surface justifies far tighter limits than an internal mounting bracket.
Process Controls That Prevent Cosmetic Rejects
Prevention beats sorting. The controls that pay back fastest in our Dongguan plant:
- Incoming coil inspection. Record surface condition, grain size, temper, and coating weight per lot. Reject at the door, not at the press.
- Die maintenance schedule based on strokes, not calendar. Polish and inspect strip-contact surfaces at a defined stroke interval.
- Slug and scrap audits. Confirm every station clears on every stroke. A single trapped slug can scrap thousands of parts.
- Lubricant control. Concentration, pH, and contamination checks. Dirty lubricant carries fines that scribe surfaces.
- First-off and periodic cosmetic checks under fixed lighting. Same lamp, same distance, same operator training.
- Handling discipline. Gloves, layer separators, no bulk dumping for cosmetic or plated parts.
When cosmetic problems persist despite these controls, the cause is often in the die itself — clearance, alignment, or wear. Our stamping die troubleshooting checklist walks through the diagnostic sequence station by station.
Frequently Asked Questions
Q: What is the difference between a scratch and a roll mark on a stamped part?
A: A scratch displaces material — under magnification you see a raised lip and often torn metal at the edges. A roll mark is a smooth depression or band with no displaced material, transferred by rolling contact. This matters because the fixes differ completely: scratches point to die burrs, clearance, or scrap dragging, while roll marks point to the coil mill, feed rolls, or guide alignment.
Q: Can orange peel be removed by polishing the die?
A: Rarely. Orange peel originates in the material's grain structure and strain distribution, not in die surface finish. Polishing a die radius reduces friction and helps marginally, but if the grain size is coarse or the temper too hard, the surface will still roughen. The reliable fixes are specifying finer grain, a softer temper, or a larger inner bend radius.
Q: How do I know if roll marks come from the coil or from my press?
A: Inspect the strip immediately before it enters the die. If the marks are already present, they are coil-origin and no die adjustment will remove them. If the strip is clean on entry and marked on exit, the source is inside the press — feed rolls, guides, stripper plate, or draw beads. A twenty-minute trial run settles the question.
Q: What cosmetic tolerance is realistic for production stamping?
A: For general industrial parts, limits such as scratches no deeper than 10% of material thickness, no scratch over 5 mm on visible faces, and no orange peel visible at 300 mm under 500 lux are workable and repeatable. Visible or plated surfaces need tighter limits. The critical step is agreeing a signed physical limit sample before production, not after the first shipment.
Q: Does plating hide or reveal stamping surface defects?
A: It usually reveals them. Plating is thin — typically 1–5 µm for nickel or gold flash — so it conforms to the underlying surface rather than filling it. A shallow scratch becomes a visible plating line; a rough orange-peel surface becomes a dull, uneven deposit. Contamination in a scratch can also cause adhesion failure or corrosion initiation. Fix surface condition before plating.
Related Resources
- About BQUQ and our Dongguan production setup: /about/
- Custom metal stamping capabilities and tolerances: /custom-metal-stamping/
- Stamped terminals and contacts: /stamping-terminals-contacts/
- Stamped brackets and mounts: /stamping-brackets-mounts/
- Industry trends and sourcing context: /industry-dynamics/
- Technical articles library: /bquq-blog/
- Frequently asked questions: /faq/
- Case studies: /case/
- Contact our engineering team: /contact/
Authored by the BQUQ Engineering Team. BQUQ (Dongguan) runs CNC machining (±0.005 mm), metal stamping, custom springs, and heat sink production in one ISO9001 factory. Source-direct from Dongguan, China — quote in 12 hours: sc@bquq.com | WhatsApp +86 13713157787 | www.bquq.com


