Parts, materials and finishes
How to identify the material of a part you need to reproduce
A discontinued spare, with no drawing or data sheet. Before machining anything, one question needs answering: what is this made of?
A broken or worn part arrives. The machine manufacturer no longer supplies it, takes three months or asks a price that makes no sense for a consumable. It needs reproducing.
The first reaction is usually to measure and draw. That is wrong, or at least incomplete. Dimensions can be obtained with calipers and patience. What no dimension shows is what the part is made of, and that determines whether the replacement works or creates a bigger problem than the one it was meant to solve.
Why material is not a detail
Think of a sacrificial part: a support specifically intended to be damaged before the machine is. Reproduce it in a harder material and it stops serving that purpose. Instead of being marked, it marks what is opposite it: the tool, fixture or product. You have turned a ten-euro consumable into a breakdown.
Conversely, if it is too soft, it lasts a week and the customer is stopped again.
There is a narrow range between those errors, and getting it right starts with identifying the original alloy.
The method: weight and volume
You do not need a laboratory for the first step. You need a precision scale and the CAD model you have to produce anyway.
Density is mass divided by volume. Weigh the original and know the exact volume of your model, and you have the material density. Although density does not identify a specific alloy, it eliminates entire families immediately.
The procedure is:
- Build the CAD model from measured dimensions, in as much detail as possible. Every hole, chamfer and recess counts.
- Get the volume from CAD. Any package gives it in mm³.
- Weigh the original part on a scale with at least 0.1 g resolution.
- Divide. Mass in grams divided by volume in cm³ gives density in g/cm³. Remember that 1 cm³ equals 1,000 mm³.
A real case
Some backing pads from a card chip-insertion machine — known in the industry as chip embedding or implanting. These are the sacrificial supports underneath while the cutter opens the module cavity.
They arrived without a drawing or material reference. They looked like bronze, but “bronze” is not a material: it is a family with dozens of alloys and very different behaviours.
These were the numbers:
| Measurement | Value |
|---|---|
| Original part mass | 54 g |
| CAD model volume | 6,034.11 mm³ |
| Volume in cm³ | 6.034 cm³ |
| Resulting density | ≈ 8.95 g/cm³ |
With that number, the table does the work:
| Material | Approximate density (g/cm³) |
|---|---|
| Aluminium alloys | 2.70 |
| Titanium | 4.50 |
| Zamak | 6.60 |
| Grey cast iron | 7.00 – 7.30 |
| Steel | 7.85 |
| Stainless steel 304 / 316 | 7.90 – 8.00 |
| Brass CuZn39Pb3 | 8.45 |
| Leaded tin bronzes | 8.70 – 8.90 |
| Pure copper | 8.96 |
| Lead | 11.34 |
Aluminium, zamak, cast iron, steel and stainless are ruled out without question. Brass falls short. We are in the range of leaded tin bronzes or practically pure copper.
Pure copper is ruled out by function: it is too soft and too expensive for a consumable, and does not machine well. That leaves bronzes of the CuSn7Zn4Pb7-C type — listed in English-language catalogues as leaded gunmetal or SAE 660. A leaded bronze designed specifically for easy machining and predictable wear. Exactly what a sacrificial part needs.
What this method does not tell you
This is where honesty matters, because density is not a magic wand.
It does not distinguish neighbouring alloys. Several bronzes fall within the same tenth. Density gets you to the correct family; choosing the exact member requires other criteria.
The example result came out high. Nominal CuSn7Zn4Pb7-C is around 8.80 g/cm³, while the measurement gave 8.95. A 1.7% difference. For a 54-gram part that can be explained by scale tolerance, small differences between the CAD model and the real part, or a slightly different composition in the original batch. It is a normal deviation, but it must be noticed rather than ignored.
A worn part weighs less than a new one. Measuring a heavily worn part against an unworn model volume gives a low density. Use the least damaged part in the batch, or model the wear.
Coatings lie. Chrome plating, anodising or paint affect weight and volume in ways that are difficult to estimate.
How to complete the identification
Density is the first filter, not the verdict. Afterwards:
- The colour of chips and a fresh cut. Freshly machined bronze has a characteristic tone, different from brass.
- Machining behaviour. Leaded bronze produces short chips and cuts with a recognisable smoothness. Lead-free bronze sticks and drags.
- Function in the machine. What the part is expected to do narrows the options more than any measurement.
- Spectrometry, if order volume or criticality warrants it. It is quick and not always expensive.
- Ask. Sometimes the machine manufacturer will tell you the material even if they will not sell the part.
In summary
Material identification is the first half-hour of any reproduction job, and the one that most affects the outcome. Weight, volume and a density table solve most of the problem without leaving the workshop or spending a euro.
The rest is craft: knowing what the part has to do inside the machine and choosing accordingly.
At 7Axis Industrial we manufacture parts, spares and short runs for production machinery, including parts the manufacturer no longer supplies. We regularly work with chip-insertion equipment, card personalisation and printing lines.
View the full reversible backing-pad case → Parts, spares and short runs →