Copper vs aluminium busbar, from the machine operator's side
Not the procurement comparison. What actually changes on the punch, the bender and the deburrer when you switch between copper and aluminium bar.
10 min readUpdated 2026-08-18
If you came here to work out whether to specify copper or aluminium busbar, this is the wrong article. That decision is made on conductivity, weight, price per kilogram, short-circuit withstand and installed cost, it has been written about exhaustively, and you will find better treatments of it than anything we could add. Go and read one of those.
This is about what happens after that decision has been made and a pallet of the other metal turns up in your shop. The punch settings are wrong, the bend angles are wrong, the deburring line loads up, and the first joint you make will be wrong too. None of that is in the procurement comparison.
Start with the mechanical numbers, not the electrical ones
Everything downstream comes from these. Typical values for the two standard busbar materials, high-conductivity copper and 1350-grade aluminium:
| Property at 20 °C | Copper C101 | Aluminium 1350 | Units |
|---|---|---|---|
| Tensile strength, annealed | 200–250 | 50–60 | N/mm² |
| Tensile strength, half hard | 260–300 | 85–100 | N/mm² |
| 0.2% proof strength, annealed | 50–55 | 20–30 | N/mm² |
| 0.2% proof strength, half hard | 170–200 | 60–65 | N/mm² |
| Elastic modulus | 116–130 | 70 | kN/mm² |
| Coefficient of expansion | 17 × 10⁻⁶ | 23 × 10⁻⁶ | per K |
| Melting point | 1083 | 660 | °C |
Two things to hold onto. Aluminium is roughly a third the strength of copper at the same temper, and it is roughly half the stiffness. Those two ratios are different, and the difference is what makes springback behave counterintuitively.
One more caution before we go further. "Aluminium busbar" is not one material. 1350 in the table above is the soft, high-conductivity grade. A lot of European and Asian busbar is EN AW-6101B in T6, which has a minimum 0.2% proof strength of 160 N/mm² and a minimum tensile strength of 215 N/mm², at ≥52% IACS conductivity. That is nearly triple the proof strength of half-hard 1350, and it changes every process setting in this article. Find out which one you have before you touch a machine.
Force, and the machine you already own
Punching force is perimeter × thickness × shear strength. The shear strength is where the metals separate.
Published punching data for C11000 gives about 179 N/mm² in shear for half-hard copper and 193 for hard. Soft 1xxx aluminium sits far lower: 1100-O is listed at 62 N/mm². Take a 13.5 mm hole in a 10 mm bar, so 42.4 mm of perimeter: half-hard copper needs about 76 kN, soft aluminium about 26 kN.
Be careful where those copper numbers come from. Figures in the 200 to 230 range circulate widely, but they are tensile values being used as shear values. For copper, measured shear runs at roughly 0.65 to 0.70 of tensile, not the 0.8 rule of thumb, so borrowing the tensile figure over-states the force by about a third.
Run your own hole sizes through the punching force calculator rather than scaling from that example, and pick the grade that matches what is actually on your rack. If your aluminium is 6101B-T6 rather than 1350, 62 N/mm² is badly optimistic; something nearer 150 N/mm² is the right starting assumption for tonnage, which is why the calculator lists the aluminium grades separately instead of offering one "aluminium" option.
The practical consequence for a shop already tooled for copper is that force is never the constraint on aluminium. An EMAC-BP-60 at 600 kN is heavily over-specified for soft aluminium bar, which is fine. What is not fine is assuming the process is therefore easier.
Soft does not mean easy
This is the part that catches people out. Copper is harder, but copper is better behaved.
Copper is a relatively brittle-fracturing material in punching terms. The punch penetrates, a crack initiates from the punch edge and propagates to meet the crack from the die edge, and the slug separates. The transition from the burnished band to the fracture zone is clean and the burr is small and consistent.
Soft aluminium does not want to fracture. It wants to flow. It stretches ahead of the punch, drags down into the die, and separates late and untidily. Three specific problems follow.
The first is galling and built-up edge. Aluminium transfers to the tool. Work on 2 mm AA5754-O sheet punched with M2 tool steel measured the transferred volume directly and found galling proceeding fastest in the first strokes of a dry punch, settling to a rate of about 7.5 × 10⁵ µm³ per stroke between the 20th and the 125th. Lubricating the sheet with an oil-based lubricant markedly reduced the transfer. What the transferred material does is change the effective punch geometry: the punch grows, the clearance closes, hole size drifts, and eventually the punch seizes in the slug.
The second is cold welding, which is the same adhesion acting on the slug. An aluminium slug that has welded itself to the punch face comes back up with the punch and lands somewhere it should not, usually on the die face, where the next stroke presses it into the bar.
The third is the shape of the burr. A copper burr is a raised lip you can knock off. An aluminium burr is a rolled-over ductile fin that folds under a deburring brush rather than shearing away, and stands back up when the bar flexes.
Clearance goes the other way from your instinct
The general rule is that harder, higher-shear-strength material takes more clearance and softer, more ductile material takes less. That is the opposite of what most operators guess, because they reason from tool life rather than from cut quality.
Set aluminium clearance too wide and it does not fracture cleanly. The material is dragged and stretched into the die, giving a large rollover, a conical hole that is smaller at entry than at exit, and a heavy tearing burr on the underside. Set it too tight and you get secondary shear, a visible second fracture band on the hole wall, plus the galling problem gets worse because the punch and die land are closer together with aluminium between them.
Published clearance tables for busbar do not agree with each other, and some of the ones circulating are internally inconsistent, so use the hole wall as the instrument. A clean burnished band with a single fracture zone below it is right, a double fracture band means too tight, and a large rollover with a torn exit means too wide. The numbers and the full diagnostic sit in punch-to-die clearance for copper and aluminium busbar.
The other correction that matters is punch nose geometry. A flat-faced punch that works in copper will need a shear angle or a bevel on aluminium to break the cut progressively and drop the peak force, which reduces both the snap-through shock and the tendency for the slug to stick.
Springback: it depends which aluminium
Springback in a bend scales with the ratio of yield stress to elastic modulus, σy/E, for a given bend radius to thickness ratio. Not with strength alone, and not with stiffness alone.
Work the ratio out for the three common busbar materials:
| Material | 0.2% proof | E | σy/E |
|---|---|---|---|
| Copper, half hard | ~185 N/mm² | ~123 kN/mm² | 1.50 × 10⁻³ |
| Aluminium 1350, half hard | ~62 N/mm² | 70 kN/mm² | 0.89 × 10⁻³ |
| Aluminium 6101B-T6 | 160 N/mm² min | 70 kN/mm² | 2.29 × 10⁻³ |
Half-hard 1350 springs back appreciably less than half-hard copper, because although it is much weaker it is also much less stiff and the strength drops further than the stiffness does. 6101B-T6 springs back roughly half as much again as copper, because it recovers most of copper's strength while keeping aluminium's low modulus.
This is why "aluminium springs back more" and "aluminium springs back less" are both statements you will hear from experienced people, and why both of them are right about the material they happen to work with.
The practical answer is not to argue about it but to measure it. The bending machines carry electronic springback compensation, which means the correction is a stored value per material and per tool combination rather than an operator's habit. Set up three material entries, not one called "aluminium", and bend a test coupon for each new temper or supplier. The same discipline applies to the minimum bend radius, which aluminium generally needs to be more generous with despite the lower forces, because the ductility that makes it flow under a punch does not protect it from cracking on the outer fibre of a tight bend.
Note also the 23 × 10⁻⁶ per K expansion coefficient against copper's 17. Over a 6 m bar that changes how much length you have when you measure it in a cold shop and where the holes end up when the assembly is hot.
The oxide layer changes everything downstream
Copper oxidises slowly, and the film that forms is semiconducting, because copper ions diffuse into it and conduction can proceed through it in both directions. This is why bare copper joints work.
Aluminium oxide is a first-class electrical insulator and it forms within microseconds of a fresh surface meeting air. Every aluminium bar you receive already has it. Every cut face and every punched hole grows a new one before you can put the bar down.
It is also mechanically hard. Corundum is an abrasive, with hardness in the thousands on the Vickers scale, against roughly 90 HV for hard-drawn copper. So the film that will not conduct is also the film that wears your tooling, which is an unpleasant combination: aluminium loads the tool by adhesion and abrades it by oxide at the same time.
Two consequences in the shop.
For deburring, aluminium loads abrasive belts and brushes far faster than copper does, and the loaded media then burnishes rather than cuts. Run aluminium with dedicated media. Do not share a belt between the two metals in either direction: copper particles embedded in an aluminium surface are galvanic cells waiting for humidity, and aluminium smeared into a copper joint face is an insulating film in exactly the wrong place. The BND800-2 takes bar from 0.5 to 50 mm thick at 5 to 30 m/min, so running the two metals as separate campaigns with a media change between them costs scheduling rather than throughput.
For jointing, the oxide has to be broken mechanically at assembly and then kept from reforming. Standard practice is to abrade the mating faces under a layer of jointing compound, so that fresh metal never sees air, and to assemble immediately. A bar prepared on Monday and bolted on Wednesday has been prepared for nothing.
Plating and galvanic pairing
Bare aluminium-to-aluminium joints can work if they are compounded and assembled properly, but they are much less forgiving than bare copper-to-copper, because the oxide is insulating rather than semiconducting and because aluminium creeps.
Carry that creep figure around. Annealed 1080 aluminium at 20 °C shows a minimum creep rate of 0.022% per 1000 hours at 26 N/mm², and annealed high-conductivity copper only reaches the same rate, at the same stress, when heated to 150 °C. Room-temperature aluminium under normal joint contact pressure is already creeping at a rate copper does not reach anywhere in its service range. Disc-spring washers on aluminium joints are not an upgrade, they are the baseline, and there is more on why in the bolted joint article.
Where aluminium meets copper, plating stops being optional. The two metals sit far apart electrochemically, at +0.34 V for copper and −1.66 V for aluminium against the standard hydrogen electrode, and with any electrolyte present the aluminium corrodes. The rules are short:
- Never bolt bare aluminium to bare copper anywhere that can see moisture or condensation.
- Tin plate the aluminium face so both sides of the interface are the same metal, or use a factory-bonded bimetallic transition plate.
- If you have to make the transition in the field, seal the interface with a compound and accept an inspection interval.
- Keep steel fasteners out of direct contact with aluminium too. The pairing is less severe than with copper but it is not benign, and a plated or stainless fastener with an insulating washer is cheap.
Tool wear and lubrication
Copper is normally punched dry. Copper is abrasive on tooling in the ordinary way but it does not adhere, so tool life is set by edge wear and is predictable.
Aluminium should not be punched dry if you can avoid it. Lubrication is the single most effective intervention against galling, and it does two jobs at once: it stops metal transferring to the punch, and it lets the punch withdraw from the slug without dragging. The trade-off is that lubricant on a busbar has to come off before insulation or plating, so it needs to be a grade your degreasing process can actually remove, and the cleaning step needs to be in the routing rather than assumed.
Tool coatings help. Nitrided or DLC-coated punches resist aluminium pickup better than plain tool steel. Frequency of punch inspection should also change: with copper you inspect on a stroke count, with aluminium you inspect on hole quality, because galling degrades the hole long before the punch edge is worn out.
What to change when the metal changes
If you are switching a line over, the short list is:
- Confirm the alloy and temper from the mill certificate, not the packing note. 1350 and 6101B-T6 behave differently enough to be treated as separate materials.
- Reset the shear strength in the force calculation and re-check tonnage against the hole pattern that actually gets punched most.
- Reduce clearance for soft aluminium and verify against the hole wall, not against a table.
- Introduce lubrication on the punch and a cleaning step downstream of it.
- Re-derive springback per temper by bending coupons, and store it as a separate material entry in the bender.
- Swap deburring media and keep them segregated.
- Change the joint specification: compound, immediate assembly, disc springs, and plating at any aluminium-to-copper interface.
Copper's own temper matters as much within its family, and EN 13601 is the reference for that: R220/H040 soft annealed at 220 to 260 N/mm² through to R290/H090 hard at 290 to 360 N/mm². A shop that runs one temper of copper and one grade of aluminium has two materials to manage. A shop that takes whatever the trader has in stock has considerably more than two, and the machine settings need to reflect it.