Sourcing Invar Parts as a Laser System OEM

By Invar Machining editorial team· Published September 26, 2026 · How we write these guides

A telescope program buys one Invar bench and lives or dies on how well that single part performs. A laser system OEM buys a cavity spacer or a mirror mount and needs the fiftieth one off the line to behave the same as the first, then survive a truck, a customs warehouse and a forklift before it reaches the customer. That is a different sourcing problem from the one most Invar content addresses, and it changes what you should ask a machining vendor before you commit to a production run.

What changes when you are an OEM, not a lab

A research or aerospace program usually qualifies a design once and builds a small number of units under close supervision. A laser system OEM builds the same assembly repeatedly, often with a supply chain that includes more than one operator running the machine and more than one batch of raw material over the life of the product. The part that matters is not the best one a shop can make. It is the average one, with a known and narrow spread around it.

This shows up directly in two places: the internal dimensions that set cavity length and mirror position, and the mounting interfaces that align the part to the rest of the housing. A single unit that misses nominal by a few microns is a scrap part. A production run where every unit misses nominal by the same few microns in the same direction is often fine, because the system was aligned and tested against that offset. What breaks a production line is inconsistency: some units high, some low, with no pattern the assembly or test process can absorb.

Where Invar sits in a laser system

Invar and Super Invar typically appear in three places in a laser system, covered in more general terms in Invar for optical mounts and telescope structures:

• Cavity spacers, where the length between two mirrors sets the resonator’s mode structure directly. Growth here is not a cosmetic problem, it changes the laser’s output.
• Mirror and lens mounts, where the mounting face position sets beam pointing and alignment relative to the rest of the optical path.
• Baseplates and structural frames, which hold the whole assembly’s relative geometry as the housing heats from the pump source or the environment around it.

An OEM designing a new platform typically starts with a mount or spacer qualified for a single prototype, then has to decide what has to change to make that same part reproducible across a production run. That is the qualification gap this article addresses. The machining sequence itself, rough, stress relieve, finish and inspect at 20 °C, is unchanged from a one-off part and is covered in Invar machining tolerances, and the general cutting parameters for the base alloy are on the Invar 36 machining page.

Evaluating a shop before committing to volume

A shop that produces one excellent prototype has demonstrated it can machine Invar competently. It has not yet demonstrated it can run a production batch. Before committing volume, an OEM should look past the first article and ask about the process behind it: whether the stress relief cycle is written down and fixed, whether fixturing is designed for repeat setups rather than a one-off clamp job and whether the shop has run a batch of any tightly toleranced part before, in Invar or another material. A shop that answers these questions with specifics, rather than “we’ll figure it out,” is a better bet for volume even if its prototype quote was not the cheapest.

Typical repeatability, not just single-part tolerance

A drawing tolerance tells a shop what one part must hit. It says nothing about how tightly a batch clusters. The table below gives typical repeatability figures, the spread a well-controlled shop can hold across a run, next to a typical single-part tolerance for comparison. Both are handbook-style planning numbers, not a guarantee for any specific shop or drawing.

Feature	Typical single-part tolerance	Typical part-to-part repeatability, qualified process
Cavity spacer length	±0.010 to ±0.025 mm	±0.003 to ±0.008 mm
Mirror mount face flatness	0.005 to 0.015 mm	±0.002 to ±0.005 mm
Mounting bore position	±0.010 mm	±0.003 to ±0.006 mm
Bore diameter	H7	±0.002 to ±0.004 mm around the mean

A shop can hold the left column on a single part with a careful setup and still drift across a batch if fixturing, tool wear or the stress relief cycle is not controlled. The right column only comes from a documented, repeated process, which is what a production RFQ should be asking for, not just a single first article.

Handling and shipping stress, not just temperature

Most Invar content focuses on thermal stability because that is the alloy’s reason for existing. For an OEM shipping finished assemblies, mechanical stress from handling matters just as much and gets less attention. A spacer or mount that is dimensionally stable on the bench can still shift if it was clamped too hard during final assembly, if the housing flexes in transit, or if a drop or shock event during shipping moves a press fit or a fastener preload that the alignment depended on.

This is not a machining defect. It is a design and process interface between the machined part and the rest of the assembly. What a shop can do on its side is make sure the part itself carries no residual stress from machining that would make it more sensitive to handling loads than it needs to be, which again comes back to a proper stress relief between roughing and finishing, described in stress relief for Invar machined parts. What the OEM controls is the assembly and packaging design: consistent fastener torque, isolation from shock during shipping and a final alignment check after the unit has been through its actual shipping path, not just off the bench.

A useful discipline for a new platform is running a small number of units through the actual shipping path, whatever carrier and packaging the product will use in the field, before locking the design. Measuring alignment before and after that trip catches interface problems, a fastener that backs off, a foam insert that lets the assembly shift, that no amount of machining precision on the Invar part itself would prevent. It is a cheap test to run once and an expensive failure mode to discover from field returns.

Worked example: qualifying a spacer for a 50 unit per month run

This is a hypothetical planning example, not a real order or a real vendor’s numbers. A laser OEM is scaling a resonator platform from a hand-built prototype to a production target of 50 cavity spacers a month, each 120 mm long, with a cavity length tolerance of ±0.015 mm on the drawing.

The OEM asks its shortlisted shops for repeatability data from a 25-piece pilot batch rather than a single first article. One shop returns a mean length within 0.005 mm of nominal and a spread of ±0.004 mm across the batch, comfortably inside the ±0.015 mm drawing tolerance with margin to spare. A second shop’s single first article also lands on nominal, but the shop has no batch data and admits its stress relief cycle is “whatever the operator sets,” not a fixed, documented cycle.

The first shop’s data supports a production order with confidence, because the spread is known and the process that produced it is repeatable. The second shop might make an equally good single part, but the OEM has no basis for predicting unit 30 or unit 300, which is the actual question a production order needs answered. This is the kind of qualification gap second-source qualification addresses when a program adds a backup vendor later, and the cost curve behind moving from a prototype quantity to this kind of run is covered in prototype vs production Invar machining.

What to put on the RFQ for a production laser part

An RFQ aimed at a production laser assembly should ask for more than a quote on a drawing:

• The stress relief and any stabilization cycle, specified on the drawing or required as a fixed, documented shop process, not left open
• Repeatability data from a pilot batch, not just a single first article, when the drawing tolerance is tight relative to typical Invar tolerances
• Inspection method and temperature, normally a soak at 20 °C, and whether the shop can supply a CMM report per unit or per batch, covered in Invar CMM inspection reports
• Expected monthly or quarterly volume, so the shop can quote fixturing appropriate to a repeat run rather than a one-off setup
• Whether a second source will eventually be added, so the first shop’s process gets documented in a way a second shop could later match

Background on the family of alloys and their typical properties is on the Invar specs reference page. More on the mechanics of the low-expansion effect itself, and why it matters for a cavity that has to hold its length, is in the Wikipedia entry on the laser cavity’s governing physics.

When you send an RFQ with your drawing, we match it with up to three shops that machine the nickel-iron alloys and can speak to repeatability across a run, not just a single sample part. Each shop quotes you directly, and you compare their process and data before committing to volume. This guide sits under the optics, photonics and laser systems overview, alongside the sourcing and process guides for the rest of that vertical.

Frequently asked questions

Reviewed against

• ASTM F1684, Standard Specification for Iron-Nickel and Iron-Nickel-Cobalt Alloys for Low Thermal Expansion Applications
• Wikipedia, Laser

Values in this guide are typical handbook figures for planning. The mill certificate for your heat lot and the shop's quote for your geometry govern.

Keep reading

• Invar for Optical Benches and Telescope Structures: Why CTE Stability Matters · Materials
• Second-Source Qualification: Requalifying an Invar Part at a New Shop · Choosing a Supplier
• Prototype vs Production Invar Machining: What Changes · Cost
• Reading an Invar CMM Inspection Report: What Each Number Means · Process

All choosing a supplier guides