Invar Machining for Spectrometer and Analytical Instrument Manufacturers

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

A spectrometer measures wavelength by measuring an angle or a position, which means its optical bench is doing double duty as both a structure and a measuring instrument. If the grating mount, the monochromator arm or the detector rail moves a few microns with temperature, the instrument does not just get a little less stiff. It reports a slightly wrong wavelength, and depending on the application, that error can be the difference between a passing and a failing calibration check. This is why Invar 36 shows up inside spectrometers, monochromators and other dispersive analytical instruments even when the rest of the housing is ordinary aluminum. This guide covers where Invar fits in this instrument class, how thermal drift becomes calibration error, and what an instrument manufacturer should put on an RFQ to get parts that hold up across a production run, not just on one prototype.

From thermal expansion to a wavelength error

A diffraction grating spectrometer determines wavelength from the angle of the grating and the position of the detector relative to the incoming beam. Both of those are set mechanically, by the bench or arm the grating and detector are mounted to. If that structure expands, the grating tilts slightly or the detector shifts along its rail, and the instrument reports the wrong wavelength for a given angle, an error that shows up in the field as calibration drift rather than as anything visibly wrong with the hardware.

Because the relationship between mechanical movement and wavelength error is geometric, a spectrometer with a longer optical path or a higher dispersion grating is more sensitive to the same absolute movement than a compact, lower-resolution instrument. That is one reason high-resolution research spectrometers reach for Invar 36 more readily than a simple benchtop colorimeter, where a slightly wider calibration tolerance makes aluminum acceptable. The mechanics of how a grating disperses light by wavelength and angle, background useful for a design engineer new to this instrument class, are summarized on Wikipedia’s entry on diffraction gratings.

Instrument designers sometimes ask whether a well-characterized aluminum structure, calibrated against its own known expansion and corrected in firmware, could do the same job for less material cost. It can, up to a point, and many portable instruments do exactly that. The tradeoff is that firmware compensation only corrects for the temperature the instrument’s own sensor reports, with whatever lag and averaging that measurement carries, while a low-expansion structure simply moves less in the first place. For a research-grade instrument that has to meet a tight calibration spec without leaning on active correction, the mechanical answer is usually the more robust one, and it is the reason Invar 36 remains the default rather than a legacy choice in this hardware class.

Where Invar shows up inside the instrument

• Grating mounts and turrets, where the mounting face and the pivot the grating rotates on need to hold their geometry as the instrument warms from a cold start or an HVAC cycle.
• Monochromator arms, the structural link between the entrance slit, the grating and the exit slit or detector, where length change directly maps to a wavelength calibration shift.
• Optical benches inside the housing, the internal frame that mirrors, lenses and the grating all reference, distinct from the external housing which is rarely Invar.
• Detector mounting rails, particularly in instruments with a moving detector stage, where rail straightness and length both matter.
• Reference etalons and calibration fixtures used in the manufacturer’s own test and calibration process, not just in the shipped instrument, where drift in the fixture corrupts every unit calibrated against it.

The broader case for Invar in optical structures, including telescopes and larger benches, is covered in Invar for optical mounts and telescope structures. Spectrometers are a narrower case of the same problem, scaled to a benchtop instrument and tied specifically to a calibration number rather than a general alignment tolerance.

Typical tolerance by spectrometer class

Instrument manufacturers do not all need the same tolerance, and specifying Invar does not by itself guarantee a given wavelength accuracy. The table gives typical figures across instrument classes, for planning a drawing. The optical design sets the number that actually matters.

Instrument class	Typical mechanical tolerance on the critical mount	Typical resulting wavelength stability
Research grade high-resolution spectrometer	0.01 to 0.02 mm, arcminute angular tolerance	Sub-nanometer over a moderate temperature swing
Process or QC analytical instrument	0.02 to 0.05 mm	Fraction of a nanometer to low nanometers
Benchtop teaching or general purpose instrument	Standard machine shop tolerance	Not usually calibration critical at this level
Portable or field instrument	0.02 to 0.05 mm, plus wider operating temperature range	Depends heavily on active compensation electronics

Values are typical and vary by instrument design. A portable instrument often pairs Invar mounting hardware with active thermal compensation in firmware rather than relying on the mechanics alone, since it cannot assume a lab environment.

None of these figures substitute for the optical design’s own error budget. A tolerance stack-up that already spends most of its margin on the grating’s ruling accuracy, the detector’s pixel pitch or the entrance slit width leaves less room for mechanical drift than one where the mechanical structure is the dominant error source. An instrument manufacturer deciding between Invar 36 and a lower cost material should start from that error budget rather than from a general rule of thumb about what “precision instruments” are supposed to use.

What changes for an instrument manufacturer versus a one-off lab part

A research lab ordering one custom fixture cares about that one part. An instrument manufacturer ordering the same grating mount as a production component cares about every unit across a run behaving the same way, for two practical reasons.

Calibration is checked per unit, and outliers cost service time. If unit 30 out of a batch of 200 drifts more than the others after shipping, it usually shows up as a warranty or recalibration call, not as a machining defect anyone can point to. Consistent stress relief and stabilization across the whole run, covered in Invar 36 machining and in more depth in stress relief for Invar machined parts, reduces that spread more reliably than tightening the drawing tolerance on paper.

A second source matters for a product that stays in production for years. An instrument model that sells for five or ten years outlives a lot of shop relationships. Qualifying a second shop against the same drawing early, rather than after the first shop’s pricing or capacity changes, is covered in second-source qualification for Invar parts and is worth doing before it becomes urgent.

A hypothetical worked example: a monochromator arm at 18 to 24 °C

Take a hypothetical monochromator with a 250 mm arm between the grating pivot and the exit slit, specified to hold wavelength calibration within 0.5 nm across a lab environment that swings from 18 to 24 °C, a 6 °C range.

In 6061 aluminum, at a typical CTE of 23.5 ppm/°C, the arm grows by 250 mm × 23.5 ppm/°C × 6 °C, about 0.035 mm over the swing. Depending on the instrument’s dispersion, a movement of that size at the exit slit can correspond to a wavelength shift well above the 0.5 nm target, which is why aluminum arms in high-resolution instruments usually need either a much wider calibration tolerance or active compensation.

In Invar 36, at a typical 1.2 to 1.6 ppm/°C, the same arm grows by 0.0018 to 0.0024 mm over the same swing, roughly a fifteenth of the aluminum figure. Translated through the same optical geometry, that keeps the resulting wavelength shift comfortably inside the 0.5 nm target without relying on firmware compensation to do the whole job.

This example is illustrative. The actual relationship between mechanical movement and wavelength error depends on the specific optical design, and a real instrument’s tolerance budget should come from the optical engineer, not from this calculation alone.

What to put on the RFQ

• The specific wavelength stability target and the temperature range it applies over, so the shop understands what the mechanical tolerance is actually protecting
• Angular tolerance on the grating mount or pivot, in addition to linear dimensions
• Production quantity and whether this is a prototype run or an established product line, since that changes how the shop plans material and stress relief batching
• Surface finish requirements on optical mounting faces, covered in general terms in Invar machining tolerances
• Any thermal cycling or calibration acceptance test the finished part needs to pass, discussed in Invar thermal cycling test

See our industry overview for scientific instruments and research labs for related hardware categories, and the Invar specs reference for the specification numbers to put on the drawing.

Getting a quote

Spectrometer and analytical instrument hardware rewards a shop that treats consistency across a production run as seriously as any single part’s tolerance. Send your drawing through our RFQ form with your calibration target and production quantity noted, and we match it with up to three shops that machine Invar to the tolerance and repeatability this instrument class needs.

Frequently asked questions

Reviewed against

• ASTM F1684, Iron-Nickel and Iron-Nickel-Cobalt Alloys for Low Thermal Expansion Applications
• Diffraction grating, Wikipedia

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
• Invar Machining Tolerances: What a Shop Can Hold and What It Costs · Process
• Specifying a dimensional stability test for Invar parts · Choosing a Supplier
• Second-Source Qualification: Requalifying an Invar Part at a New Shop · Choosing a Supplier

All materials guides