Punch-to-die clearance for copper and aluminium busbar

Clearance is the gap between the cutting edge of the punch and the cutting edge of the die. It is a single number, it is easy to get wrong, and on busbar it is wrong more often than any other tooling parameter. Most published guidance was written for sheet stamping at 1-3 mm; busbar is 6-20 mm of soft, ductile, work-hardening copper. The percentages transfer. The consequences do not.

Two conventions exist and they differ by a factor of two. Total clearance is the difference between the die opening and the punch dimension. Per side clearance is half of that, the radial gap at any one point around the perimeter. A supplier chart saying "copper, 20%" and a shop running 10% may be describing identical tooling. Establish which convention a number uses before comparing it with anything. This article uses per side throughout, and converts quoted sources where needed.

What the cut edge is telling you

A punched hole in busbar has four zones, visible on the bore and mirrored on the slug. Reading them is the fastest diagnostic you have, and it costs nothing.

Rollover is at the punch entry face, where the material yields and draws into the hole before it cuts, leaving a rounded lip. Rollover depth grows with clearance, tool dullness and material softness, so it is not a defect on its own. On annealed copper it is always visible.

Burnish is the bright band below the rollover, where the hole wall was rubbed by the punch flank. It is smooth and straight, and dimensionally it is the hole. Dayton Lamina's tooling guidance describes optimum clearance as producing a slug with a burnished land of approximately one third of material thickness and an even fracture plane in line with it, which works as an acceptance rule for busbar too.

Fracture is the dull, coarse remainder. It runs at an angle because the crack travels from the punch edge to the die edge, and clearance sets that angle. A fracture surface that is straight, uniform and continuous with the burnish band means the two cracks met. That is the whole objective.

Burr is at the punch exit face: the last ligament of material elongating and tearing rather than shearing. Burr height rises with clearance, with edge wear and with softness.

Two failure signatures matter more than the rest. A secondary shear band, meaning a second burnished ring part way down the wall with a step between the two, says the crack from the punch and the crack from the die did not meet on the same plane and the material between them had to be sheared twice. That is too little clearance. A large rollover with a thin, ragged burr and a very short burnish band is too much.

Keep a slug from each new tooling set, labelled with material, thickness and clearance. Comparing a suspect slug against a known-good one takes fifteen seconds and settles most arguments about whether a tool needs regrinding or the clearance is wrong.

Clearance as a percentage of thickness, per side

The working expression is:

c = k × t          (clearance per side, mm)
die opening = punch size + 2c

where t is material thickness and k is the clearance ratio. For copper busbar, k sits in a 5-12% band per side. Below 5% you are into secondary shear and heat damage. Above about 12% on thick bar, rollover and slug control start to dominate. Around 10% per side is the common working figure, and it is where most tooling suppliers land for half-hard bar.

The band, worked out for the thicknesses that turn up in switchgear:

So 6 mm copper wants roughly 0.3-0.7 mm per side, and 10 mm copper roughly 0.8-1.2 mm per side working the upper half of the band. Those die openings look alarmingly large next to the punch if your instinct comes from thin-gauge work. They are correct. The die for a 13 mm punch in 10 mm copper at 10% per side is 15 mm, and the hole it makes will measure close to 13 mm, because the burnish band sets the bore, not the die.

Published tables vary, for a reason. Conic's tooling guide gives copper a clearance ratio of 0.20-0.25 as a total figure on servo and hydraulic presses, which is 10-12.5% per side, with a note to multiply by 1.4 above 3.2 mm thickness. Taken literally that pushes thick copper to 14-17% per side. UniPunch lists copper at 20% of thickness for half-hard and 25% for hard temper, again as total. Dayton's engineered-clearance work, which is thin-gauge stamping, goes higher still. The spread is real rather than sloppy: the optimum depends on whether you are buying tool life, minimum burr or dimensional repeatability, and those three do not peak at the same clearance.

Too little clearance

The crack from the punch and the crack from the die each start at their own cutting edge and propagate toward each other. With correct clearance they lie on the same plane and join. With insufficient clearance they run past one another, and the ligament between them has to be torn through in a second event. That gives the secondary shear band, a rough and inconsistent bore, and a hole undersize relative to the punch.

The consequences at the tool are worse than the consequences at the part. Copper springs back onto the punch flank after snap-through, so at tight clearance the hole closes on the punch and every stroke ends in an interference fit that has to be forced out on the return. Stripping force climbs, the flank abrades, and frictional heat builds at the tip. Dayton's field examples show discoloration behind the tip from exactly this mechanism, with the punch's heat treatment damaged and its life ended well before the edge itself is worn. In severe cases the tip fatigues and pulls off.

Withdrawal, not the cutting stroke, is where much of the wear happens. Dayton puts it at up to two thirds of total punch wear, and it is entirely a function of how tightly the hole grips the flank. That is the strongest argument against running clearance at the bottom of the band on copper. Copper is soft, so the press never complains. The tooling account does.

Peak tonnage rises at tight clearance too, because more of the thickness is sheared rather than fractured. On a machine with little headroom, that is a plausible route to a stalled ram on the thickest bar in the job.

Too much clearance

The material bends into the die before it cuts. Rollover grows, the burnish band shortens, the fracture zone takes over most of the wall and burr height rises. The hole is oversize and its edge is no longer square, which matters where the hole is a bolted joint face.

The operational failure is slug pulling. As clearance increases, the slug becomes a looser fit in the die and comes back up with the punch at withdrawal instead of dropping. It lands on the bar and gets driven into the next hole, or wedges in the stripper. On copper the mechanisms are oil film adhesion and the vacuum formed between the bowed slug and the punch face. Magnetic adhesion, the usual third culprit on steel work, does not apply.

The countermeasures are mechanical: a spring-loaded ejector pin in the punch face with a side vent hole, a shear angle that bows the slug so its own springback releases it, or a die with a short cutting land followed by a narrower relief that grips the slug so it cannot climb. Restricting punch entry into the die to about 0.5-0.8 mm at bottom of stroke helps, since deep entry creates more vacuum on the way out. The full list and what each is worth is in tool life and cost per hole.

Small holes in thick bar

Busbar has a geometry problem that thin-gauge stamping does not. An M12 clearance hole is 13 or 13.5 mm. In 10 mm bar the ratio of hole diameter to thickness is about 1.3; in 12 mm bar it is about 1.1. That is a stubby, deep hole, and it does not behave like a 25 mm hole in 3 mm sheet.

Below a diameter-to-thickness ratio of about 1.5 the slug becomes hard to bend and break free. Punch loading rises, the burnish band lengthens, the burr grows and the hole comes out undersize. Dayton's guidance is to add 1% per side to the nominal clearance once the hole drops to 1.5 times thickness, rising to roughly 4% per side extra when hole diameter equals thickness. Applied to a 13 mm hole in 12 mm copper that correction is not optional, and it explains why a tooling set that works beautifully at 6 mm misbehaves at 12 mm on the same nominal percentage.

PASS Stanztechnik's tool-life data agrees from the other direction: their derating factors are 0.6-0.8 for a cut feature smaller than 1.5 times sheet thickness and 0.3-0.5 for one smaller than 1.0 times thickness. A small hole in thick bar costs between half and two thirds of the tool life you would otherwise get.

Punch force, and what the point geometry does to it

The force required is the area being sheared multiplied by the shear strength of the material, F = perimeter × t × τ, or F = π × d × t × τ for a round hole. Worked examples and a reference set of shear strengths are in the companion article on sizing press force for busbar punching, and the arithmetic runs directly in the busbar punching force calculator.

Clearance interacts with this, and so does the point geometry. A flat-faced punch cuts the entire perimeter at once, so the whole force arrives as a spike and the snap-through unloading is violent. Grinding shear onto the punch face staggers the cut and peak tonnage falls, but on thick busbar it falls less than the marketing suggests. H. Weiss's shear-factor chart gives, on 1/4 in (6.35 mm) stock, a factor of 0.90 for a 1/16 in (1.6 mm) shear depth, 0.85 for 3/32 in (2.4 mm) and 0.65 for 3/16 in (4.8 mm). A shallow shear on thick copper buys 10%. Getting 35% needs a shear depth approaching three quarters of the bar thickness, which is a long, fragile point.

Flat gives the highest peak force, the squarest hole and the flattest slug. Single shear, a face ground at an angle across the punch, gives the largest force reduction but pushes the punch sideways, and on a stubby busbar punch that side load goes into the guide and the stripper. Rooftop, symmetrical about the centreline, balances the side loads and is the usual choice for anything but the smallest holes. On piercing, put the shear on the punch rather than the die: whichever member carries the angle distorts its side of the cut, and on a pierced hole the slug is scrap while the bar is the product.

Aluminium is not copper, and 6101 is not 1350

Aluminium changes two things. Shear strength drops a long way, and it does not drop uniformly across the grades used for busbar.

Soft 1xxx conductor grades sit near 60-80 N/mm² in shear; UniPunch lists 1100-O at 62 N/mm² and 1100-H14 at 76 N/mm², which is the right order for 1350-H111 bar. Alloyed and heat-treated bar is a different material. Published shear strength for 6101-T6 is about 22 ksi, roughly 150 N/mm², and 6061-T6 is listed at 207 N/mm². Sizing a press or a tool set for "aluminium, 80 N/mm²" and then being handed 6101-T6 bar nearly doubles the force you actually need. Establish the alloy and temper before the tonnage calculation.

Clearance for aluminium busbar runs slightly below copper at the same thickness, typically around 8-10% per side, because the lower shear strength lets the fracture propagate readily. The aluminium problem is adhesion. It galls onto tool steel, builds up on the punch flank, then scores the next hundred holes. Coating helps, and PASS Stanztechnik quote a tool-life factor of 2.0-4.0 for an appropriate coating, meaning an aluminium-specific grade rather than general-purpose TiN. Lubrication helps more. PASS's derating factor for running without sheet lubrication is 0.4-0.6, so dry-punching aluminium busbar is a false economy.

Copper temper matters the same way. Under EN 13601, Cu-ETP bar is supplied as R220 soft annealed at 220-260 N/mm² tensile, R240 half-hard at 240-300 and R290 hard at 290-360. That is far enough apart that a tool set trimmed for one temper will not be optimal on another.

Specifying burr on the drawing

Most busbar drawings say nothing about burr, and then inspection rejects parts against an unwritten standard. The mechanism for saying it properly is ISO 13715, Technical product documentation — Edges of undefined shape — Indication and dimensioning, third edition 2017.

The standard gives a graphical symbol placed on the edge, with a sign and a value inside it. The sign carries the meaning:

• + indicates permitted excess material, which the standard calls passing. A burr is explicitly defined as a special case of external passing. A single positive value means the other limit is zero, so undercut is not permitted.
• − indicates required material removal, an undercut. A single negative value means passing is not permitted, so no burr at all is allowed and the edge must be broken.
• ± indicates that either is permitted, and it may only be used with a size value.

The size is the maximum permitted deviation, written after the sign. Upper and lower limits can both be given, stacked. ISO 13715 also provides for indicating the direction in which passing or undercut is permitted, which is the clause that matters most on busbar, because a burr on the joint face of a bolted connection is a different problem from a burr on the free face.

What ISO 13715 does not do is tell you what number to write. That comes from the function of the edge:

• Bolted joint faces: the burr sits between two mating conductors and holds them apart, reducing contact area and raising joint resistance. Since IEC 61439 sets a 70 K rise limit at terminals for connection of external conductors, joint resistance is a compliance issue rather than a cosmetic one. Specify the burr side, and specify it tight.
• Hole bores where a bolt shank passes: burr here interferes with assembly and can score the bolt. Moderate control is enough.
• Edges near an insulating barrier or an air gap: an IEC 60664-1 clearance and creepage question, since a sharp raised burr locally raises field strength. It is also where somebody usually claims a standard mandates chamfering. None does. Edge treatment is a means of meeting the dielectric provisions, not a requirement in its own right.

For hole patterns to DIN 43673-1, which fixes hole location and size plus the base data for the screw connections on rectangular busbars, the burr specification belongs on the same drawing as the pattern.

Where the limit is tight enough that punching alone will not hold it, the answer is a downstream operation rather than an ever-tighter clearance. Brush or belt deburring on a machine such as the BND800-2, which handles bar from 0.5 to 50 mm thick and 100 to 800 mm wide, removes the argument. Chasing a 0.05 mm burr by tightening clearance costs punch life and hole quality at the same time.

Setting and checking it on the machine

Clearance belongs to the tool set, not the press, so on a punching and shearing machine it is fixed when the punch and die are ordered and verified when they are fitted. Measure punch and die separately and compute the clearance rather than trusting the label; a punch ground on the flank rather than the face is no longer the size stamped on it. Then check concentricity, because a per-side figure has to hold all the way round. A punch 0.1 mm off centre in a 0.8 mm nominal clearance gives 0.7 mm on one side and 0.9 mm on the other, and the burr will be visibly asymmetric. On a processing centre with a multi-station turret holding ±0.05 mm positioning, station alignment and turret cleanliness are part of the clearance chain.

What this local plastic flow does to bar straightness is dealt with in flatness, straightness and twist after processing, and the punching process overview covers where the hole sits in the rest of the route.