Lead Frame Stamping: Surface Mount and QFN Parts

A 0.4 mm pitch QFN lands on your reflow line with one lead lifted 0.08 mm out of plane, the paste wets nine pads out of ten, and the part passes electrical test but fails in the field at 90 °C thermal cycling. That is not a soldering problem. It is a lead frame problem that your stencil and profile cannot fix. Coplanarity outside ±0.05 mm, plating that lets the copper oxidize before the reel reaches the pick-up, and burrs that shadow the plated surface are the three defects that quietly convert a 0.3 cent stamped part into a six-figure sorting bill. This guide covers what a precision lead frame stamper actually controls: strip alloy, plating system, die tolerances, and the inspection data you should demand before you sign a purchase order. Along the way we compare stamping against chemical etching, walk the die design details that decide coplanarity, and lay out the cost structure so you can read a quote instead of guessing at it.

Key Takeaways

QFN lead coplanarity is specified at ±0.05 mm (JEDEC MO-220); anything wider creates open joints you cannot screen out with AOI.
Strip choice is a conductivity vs strength trade: C19400 at ~65% IACS, C7025 at ~40% IACS with ~600 MPa yield, Alloy 42 at ~4.5 ppm/°C for CTE-matched ceramic packages.
Silver plating 2-8 µm or NiPdAu pre-plated frames (Ni 1-2 µm / Pd 0.02-0.06 µm / Au flash) decide solderability shelf life: bare copper is marginal inside days, plated frames hold 6-12 months.
Demand burr ≤10% of strip thickness and die clearance 5-8% per side; both are measurable on a 2D optical comparator before you approve the first article.
Stamp the frame above roughly 0.3 mm pitch; below that, or for ultra-fine internal lead frames, chemical etching or a stamped-plus-half-etched hybrid takes over.
Qualify on first-article coplanarity Cpk ≥ 1.33 measured across the full strip width on all leads, not on three hand-picked parts.

What Lead Frame Stamping Actually Produces

A lead frame is a stamped strip that carries the die and routes its signals to the board. In surface-mount packages - QFN (quad flat no-lead), SON, DFN - the leads are coplanar pads on the package underside, and the center of the frame is an exposed thermal pad (ePad) that conducts heat into the PCB. The frame starts as strip 0.10-0.25 mm thick, typically 100-250 µm for QFN families, and is formed in a progressive die at high speed. Between the stamping stations the frame carries tie bars and dam bars that hold the leads in position; these are trimmed after molding. For fine-pitch geometries, stampers combine mechanical forming with half-etching - chemical milling that thins selected zones to roughly 50% of strip thickness so that tie bars can be removed without shearing the plated surface.

What separates a precision stamper from a general metal-stamping shop is not the press. It is the die design: 0.4 mm pitch means lead width around 0.20 mm with 0.20 mm gaps, and every one of those edges must be burr-controlled, plated, and flat enough to sit on paste. Die clearance for copper alloys runs 5-8% of material thickness per side - for 0.15 mm strip that is 0.008-0.012 mm per side. At that scale, die wear is measured in microns and tool steel grade, punch geometry, and lubrication all show up in the coplanarity data.

The Die Layout at 0.4 mm Pitch

A lead frame die is a study in controlled release. The strip advances on a carrier with pilot holes; each station does one job - pierce, trim, coin, bend, half-etch mask, and finally cut-off - so that no single operation carries the whole tolerance. Tie bars connect the leads to the carrier and each other during stamping and molding, and dam bars seal the mold cavity during encapsulation; both are removed in a trim-and-form step that must not drag burrs across the plated surface. The order of those stations, and the clearance at each, is what the first article report really documents.

What a Precision Stamper Runs

The equipment side matters less than the die side, but it still sets the ceiling. High-speed stamping of lead frame strip needs fast, repeatable feeding because a 0.4 mm pitch leaves no room for feed error. Our press fleet runs 21 machines, including Aida high-speed presses up to 300 SPM and 80 tons with ±0.005 mm feed positioning, plus dedicated EV terminal presses for the finest work. The 0.10-0.25 mm lead frame strip sits comfortably inside our 0.05-3.0 mm material range, and strip widths up to 650 mm allow multi-row layouts that spread the die cost across several frames per stroke. The tool room builds and maintains the dies in house with wire EDM and CNC, so a worn punch at 0.4 mm pitch is corrected in days, not in a vendor queue. See the progressive die high speed stamping service page and the stamped lead frames product page for the full capability list.

Stamping vs Etching: Choosing the Strip Process

Not every lead frame is stamped, and pretending otherwise gets expensive. Chemical etching cuts frames from pre-plated or bare strip with photoresist, and it wins on two axes: pitch and tooling cost. Below roughly 0.3 mm pitch, or when the frame has complex internal geometries that a die cannot reach, etching is often the only practical route. The trade is the edge profile - etched edges are tapered and rougher than sheared ones, which hurts solder fillet formation and wire bond placement - and the unit cost, which stays flat because etching is slow per frame.

Attribute	Stamping (progressive die)	Chemical etching
Finest practical pitch	~0.3 mm with tight die control	~0.15-0.2 mm
Edge profile	Sheared, square, burr-controlled	Tapered, slightly rough
Tooling cost	High (multi-station die)	Low (phototool)
Unit cost at high volume	Lowest	Higher, flat with volume
Mechanical strength	Full grain structure	Slightly reduced at edges
Typical use	QFN, SON, DFN, SOP frames	Ultra-fine pitch, internal frames, quick turns

The decision rule I hand buyers is simple. If the pitch is 0.4 mm or above and the volume is above a few hundred thousand pieces a year, stamping wins on cost and solderability. If the pitch is 0.3 mm or below, or the part needs features no die can form, etching wins and the extra cost per frame is the price of the geometry. Between those, a hybrid - stamped frame with half-etched tie bars - gives the best of both: sheared lead edges for solderability and etched release for fine internal features. Our lead frame etching vs stamping guide goes through the full comparison with cost curves and edge-profile photos.

Strip Material Is the First Solderability Decision

The alloy determines electrical performance, heat path, formability, and - critically - how the part behaves during reflow. For QFN and other SMT lead frames, three materials cover almost the whole application map.

Alloy	Conductivity	Yield Strength (typical)	CTE (ppm/°C)	Where It Earns Its Place
C19400 (CuFe2P)	~60-65% IACS	410-550 MPa (HSM-XH tempers)	~17	Workhorse QFN/SOP frames; strength + conductivity balance for mid pin counts.
C7025 (CuNi3SiMg)	~40% IACS	~600-700 MPa (HT)	~17	High pin-count, fine-pitch QFN; best stress-relaxation resistance, holds coplanarity under heat.
Alloy 42 (Fe-42Ni)	~3% IACS	~350-500 MPa	~4.5	Glass-to-metal and ceramic packages where CTE must track the die; thermal path goes through the ePad, not the leads.

Rule of thumb from thirty years of running these dies: if the package dissipates power through the leads, use copper. If the board routes high current through the frame, C19400 wins on conductivity. If pin count is high, pitch is 0.4 mm or below, and the part must survive thermal cycling without the leads creeping out of plane, C7025 is worth the conductivity penalty. Alloy 42 is not a solderability material - it is a CTE material, and its low conductivity is accepted only when the package design needs it.

Temper Sensitivity

The temper is half the alloy. C19400 ships in tempers from HSM (soft, highly formable) to XH (extra hard), and the yield spread between them is roughly 140 MPa - enough to change how the leads spring back after trimming and how they hold coplanarity through reflow. A material certificate that names the alloy but not the temper is a certificate that tells you nothing. Require the full designation (for example C19400 HSM or C19400 XH) on the RFQ and on the mill cert, and match the temper to the job: softer tempers for deep forming, harder tempers for fine-pitch frames that must resist creep.

Thickness and Width Windows

QFN frames run 0.10-0.25 mm strip, with the thinner end reserved for fine-pitch and low-profile bodies. Strip width sets the multi-up economics: a 100-200 mm wide strip can carry two to eight rows of frames side by side, and the row count is the single biggest lever on unit cost. Our presses feed strip up to 650 mm wide, which is far beyond what a lead frame needs but means a multi-row layout is never limited by the press. Ask the stamper for the planned row count before comparing quotes - a two-row layout at 300 SPM is a different price per frame than a four-row layout at the same speed.

Plating: Silver vs NiPdAu Pre-Plated Frames

Bare copper oxidizes. At 25 °C and 50% RH, a clean C19400 surface develops Cu2O within hours, and solderability degrades measurably inside days of exposure. That is why lead frames are plated before the strip leaves the die - the plating is the solderable surface, and its thickness and porosity decide how long the reel stays usable.

Finish	Typical Thickness	Why It Is Used	Shelf Life (controlled storage)
Silver (electroplated)	2-8 µm full or spot	Best solderability and conductivity; standard for QFN ePads and power packages.	6-12 months
NiPdAu (pre-plated frame, PPF)	Ni 1-2 µm / Pd 0.02-0.06 µm / Au 0.003-0.012 µm	No plating after molding; Pd blocks copper migration, Au flash protects Pd; eliminates solder-dipping steps.	12 months+
NiPdAu-Ag	Same Ni/Pd base, Ag top flash	PPF economics with silver's wire-bond compatibility.	12 months+

Two things to check on the incoming certificate. First, thickness is measured by XRF at defined points - typically a minimum of three locations per lead across the strip - and the spec is a minimum, not an average. Second, plating coverage at the lead edges: a burr that exceeds 10% of strip thickness shadows the plating bath, leaves bare copper on the shear face, and that bare edge is exactly where fillet wetting fails in AOI. If your assembler reports intermittent opens on the same lead position across lots, pull the XRF map and the burr report before you touch the profile.

XRF Verification Protocol

Write the measurement plan into the purchase order. The XRF map should name the measurement points per lead, the number of leads sampled per reel, and the acceptance criterion as a minimum thickness, not an average. For a spot-plated frame, the spot position tolerance matters as much as the thickness: a silver spot that drifts 0.1 mm off the ePad center leaves half the pad underplated. Add a porosity or edge-coverage check on the shear faces - a simple microsection of three leads per lot tells you whether the burr is shadowing the bath.

Shelf Life and Storage Discipline

Plating buys you a window, not forever. Silver-plated frames run 6-12 months in controlled storage; NiPdAu frames hold 12 months and beyond. The window is only real if the reel is stored in a nitrogen cabinet or moisture-barrier bag with desiccant, and FIFO is enforced on the plating date. A reel that sits two years in a warehouse is a future intermittent-open problem no profile change can fix.

Selective Reel-to-Reel Plating

Full coverage is wasteful and sometimes wrong. Selective reel-to-reel plating - masking everything but the bonding zones and the solderable lead tips - cuts precious-metal consumption dramatically and is the standard way to plate fine-pitch frames without bridging adjacent leads. Our plating line runs selective reel-to-reel gold, silver, tin, and nickel at 2-8 µm, which covers the silver and PPF-family finishes this section describes. The plating and surface finish guide and the terminal plating selection guide go deeper into when each finish wins.

The Tolerances Your Reflow Line Actually Feels

IPC-A-610 Classes 2 and 3 accept different fillet conditions, but no inspection class can recover a lead that is physically out of position. The geometry budget that matters, in order:

Coplanarity: all leads and the ePad within ±0.05 mm of a common plane (JEDEC MO-220 for QFN). One lead lifted 0.05 mm at 0.4 mm pitch means the paste under it is compressed to near zero; lifted 0.08 mm, the joint is open regardless of profile.
Pitch and lead position: 0.4-0.5 mm standard for QFN; position tolerance on a stamped frame is typically ±0.025-0.05 mm, held by die accuracy, not by inspection.
Twist and bow: strip-level flatness measured across the whole frame; a frame that is flat on the bench but bows 0.1 mm in reflow will wick solder unevenly on the ePad.
Burr: ≤10% of strip thickness, on the trim side, away from the plating surface.

Characteristic	Typical spec (QFN)	Measurement method
Coplanarity	±0.05 mm (MO-220)	Laser or CMM, all leads + ePad
Lead position	±0.025-0.05 mm	2D optical comparator, CMM
Pitch	0.4-0.5 mm	Optical, pitch-measuring fixture
Twist / bow	Per package drawing, often 0.05-0.1 mm	Surface plate + gauge, laser flatness
Burr height	≤10% of strip thickness	Optical comparator, microsection
Plating thickness	Minimum per spec (2-8 µm typical)	XRF at defined points

This is where IPC-2221 enters on the board side: land pattern dimensions, annular ring requirements, and conductor spacing assume the component stays inside its coplanarity envelope. The arithmetic is unforgiving. At process capability Cpk 1.0, a characteristic drifts out of spec at about 2,700 ppm; at Cpk 1.33 it drops to roughly 63 ppm. A 64-lead QFN has 64 chances per part to trip one of those limits. If your frame supplier cannot show coplanarity Cpk ≥ 1.33 on the first article report - measured on a minimum of 30 leads across the strip, not the three prettiest parts - you are buying rework, not lead frames.

From Reel to Reflow: The Assembly Interface

The frame is delivered on carrier tape per EIA-481, embossed pockets with the parts oriented for pick-up; a 5x5 mm QFN typically runs in 16 mm tape, smaller bodies in 8 or 12 mm. Three assembly-side decisions interact directly with the stamping quality:

Stencil design (IPC-7525): the ePad aperture needs an area ratio above 0.66 or paste will not release cleanly, and voiding in the thermal pad is capped at 25% by most OEM specs - automotive programs commonly pull it to 10%. Stamped ePad flatness and plating porosity feed directly into that void number.
Land pattern (IPC-2221 / IPC-7351): pads sized for the 0.4-0.5 mm pitch with a solderable fillet zone on the outside of each lead; the frame's plating must wet that zone on every side, which is why edge coverage on the shear face matters.
Storage discipline: nitrogen cabinets, moisture-barrier bags, and first-in-first-out on plating date. Silver-plated frames run 6-12 months; do not let a reel sit two years in a warehouse and then blame the profile.

One more caution specific to silver: under humidity and DC bias, silver migrates across the package surface and can bridge 0.4 mm gaps over time. If the end product lives in high-humidity, high-voltage conditions, ask for the NiPdAu or NiPdAu-Ag option before you qualify the part.

Reel-to-Reel Details That Bite

The tape pocket depth has to clear the lead tips plus the coplanarity envelope, or pick-up will drag a lead. Check the pocket depth against the frame thickness plus 0.1 mm clearance, and verify the cover tape peel strength on the reel certificate - a peel that is too aggressive lifts fine leads out of their pockets during depanel. Both are cheap to verify at incoming inspection and expensive to discover as placement rejects.

Reflow Profile Sensitivity

Lead-free profiles peak at 245-260 °C, and the frame sees that temperature at the strip level before singulation. Copper alloys soften and relax under that heat; C7025 is the alloy that holds its coplanarity best because its stress-relaxation resistance is built into the composition. If the first article coplanarity was measured on the strip and the assembly house measures it after reflow, expect a delta - the number that matters is the one after reflow, so ask the supplier whether their Cpk data is pre- or post-reflow and hold them to the same condition your line runs.

Die Design and Cost: What Drives the Quote

Clearance, Burr, and Edge Quality

Die clearance for copper alloys runs 5-8% of strip thickness per side. Too tight and the punch wears fast and the shear tears; too loose and the burr grows past the 10% acceptance and shadows the plating. At 0.15 mm strip, that is a working window of 0.008-0.012 mm per side - the difference between a clean shear face and a burr that shows up in AOI six months later. Punch material matters at this scale: carbide punches hold the clearance through long runs, and the tool room's grinding capability decides how quickly a worn station is restored to the print dimension.

Half-Etching and Tie-Bar Placement

The half-etch zone is where stamping and chemistry cooperate. Etching selected zones to roughly 50% of strip thickness lets the tie bars be removed without a shear that drags across the plated surface. The depth tolerance on that half-etch is typically ±10% of the target, and it has to be verified per lot because bath chemistry drifts. A frame with the half-etch in the wrong zone, or over-etched, loses the mechanical support the tie bars were supposed to provide during molding, and the leads shift in the mold - a coplanarity failure that shows up at the customer's AOI, not yours.

Multi-Up Economics and the Worked Example

Lead frame dies are expensive and the cost spreads across rows and strokes. A production die for a 64-lead QFN frame typically lands in the $80,000-150,000 range depending on station count, tolerance class, and the half-etch tooling; the amortization per frame collapses as volume and row count rise. The quote layers look like this:

Layer	Share of frame price (typical)	What moves it
Tooling amortization	10-30% at launch volumes	Annual volume, rows per strip
Strip material	15-25%	Alloy, temper, coil width utilization
Plating	20-40%	Finish type, spot vs full, thickness
Press time and inspection	15-30%	SPM, rows per stroke, measurement depth

The plating layer is the one buyers misread most: at 20-40% of the frame price, a silver spot-plate versus full coverage, or a NiPdAu PPF versus post-plating, is a bigger lever than the press rate. Run the comparison on total cost per million frames - die amortization plus plating plus press time - and the row count is the first number to negotiate, because doubling the rows per strip halves the press-time share of every frame. The connector terminal stamping guide covers the same cost arithmetic for terminals, and the same rules apply.

How to Qualify a Lead Frame Stamping Supplier

You are not buying a stamping; you are buying a distribution of tolerances that has to land inside your assembly window. Put these on the qualification checklist:

First-article report with coplanarity, pitch, and lead position data on a real sample size - minimum 30 parts, all leads measured, Cpk ≥ 1.33.
XRF plating thickness maps per reel, plus a porosity or edge-coverage check on the shear faces.
Burr measurement per die station, with the acceptance at ≤10% of strip thickness.
Material certificates matching the alloy and temper you specified - C19400 HSM is not C19400 XH, and the yield difference changes coplanarity under thermal load.
Die maintenance records and a change-notification commitment, because a re-sharpened die at 0.4 mm pitch drifts differently than a new one.

Ask for the measurement method too. Optical comparators and CMMs give different numbers than laser scanners; the report is only useful if the technique is repeatable across lots and across your incoming inspection.

Audit Questions

What is the current coplanarity Cpk on the hardest feature of your last three QFN programs, and at what measurement condition (strip, post-trim, or post-reflow)?
Where is the XRF map generated, how many points per lead, and what is the minimum-thickness criterion?
What is the die maintenance interval at 0.4 mm pitch, and what drifts first - clearance, burr, or coplanarity?
How do you handle an engineering change to the frame layout, and what is the die revision cycle in weeks?
Which row counts have you shipped on a 150 mm strip, and what is the typical SPM at those row counts?

Red Flags

Three answers end the conversation. A first article with coplanarity data on fewer than ten parts. An XRF certificate that reports an average thickness instead of a minimum at defined points. And a material certificate without the temper - because the temper, not the brand, is what your reflow profile actually feels.

On our side, the frame program runs on the same discipline the automotive lines run: IATF 16949:2016 certified quality system, ISO 14001:2015 environmental management, a quality lab with CMM, optical measurement, and in-line vision, and dedicated EV terminal presses for the finest pitch work. The lead frame stamping tolerances, alloys, and process guide and the electronics stamping guide cover the surrounding territory in more depth.

Your Action Roadmap

FAQ

Can a stamped frame hold 0.3 mm pitch? Yes, with tight die control, carbide tooling, and Cpk ≥ 1.33 demonstrated on the first article. Below 0.3 mm, etching or a stamped-plus-half-etched hybrid is usually the safer route, because the die clearance window at 0.15 mm strip leaves almost no room for wear.

Silver or NiPdAu for an automotive program? NiPdAu removes the silver-migration risk in high-humidity, high-voltage environments and eliminates post-mold plating. Silver stays the choice for power ePads where the lowest resistance and best solderability matter more than migration margin.

Why do my opens always appear at the same lead position? That is the signature of a plating coverage or burr problem at one die station, not a random soldering defect. Pull the XRF map and the burr report for that lead position before touching the stencil or the profile.

What must the first article report contain? Coplanarity on all leads and the ePad, pitch and lead position, twist and bow, burr per die station, XRF thickness at defined points, and the material certificate with alloy and temper - with Cpk ≥ 1.33 on the coplanarity characteristic.

How fast can prototype frames be made? Prototype soft tooling runs far faster than a production die - our rapid prototyping and pre-production services are built for exactly this stage, and the die investment only makes sense once the package design is frozen.

Next Steps

Every failed QFN joint traces back to a lead frame decision made months earlier - alloy, plating, die tolerance, or the supplier who could not show the data. Send us your package drawing with pin count, pitch, and power dissipation, and we will return a material and plating recommendation with target tolerances before you spend a cent on tooling. For a quote-ready DFM review of your lead frame, our engineering desk is one message away.