Strain gauge installs that fail in the field fail at the surface preparation, the lead transition, or the seal — the install data sheet published by HBM, Vishay, and Micro-Measurements treats these three nodes as the controlled variables, not the bond adhesive [S1][S2].
The Micro-Measurements, Vishay, and HBM technical-note consensus is that the technician must have eight pieces of data on the install drawing before the surface is touched, and an ohmmeter verification of isolation and volume resistance after placement but before the adhesive is cured [S2][S5].
Surface preparation and the tweezer handling rule
Use tweezers to take the gage from its package — fingers deposit skin oil that shifts the self-temperature-compensation number and the grid resistance at the bond line, per the practical installation notes published by strain-gauge vendors [S1]. The surface must be abraded, degreased, and conditioned with a neutral-solvent wipe (laboratory-grade acetone or isopropanol, lint-free) before the alignment jig is set. The Practical Hints for the Installation of Strain Gages manual lists installation location, grid direction, circuit diagram, cable plan, SG type, bonding method, lead material, and protective measures as the eight pieces of data the technician must have on hand before any of this begins [S2]. A small deviation in surface roughness or pH will bleed into the gauge factor and present as a zero drift on the first temperature cycle.
Ohmmeter verification before and after bonding
After alignment and before the cement is cured, the technician measures the gauge isolation and volume resistance with an ohmmeter — a faulty installation yields inaccurate measurement results, and the only way to catch a lifted grid or a solder splash at this stage is by resistance, not by eye [S5]. Micro-Measurements documents the same step in its knowledge-base technical notes covering Wheatstone-bridge shunt calibration, non-linearity, and noise control as the precondition for a valid bond. A drift between the as-received and as-bonded grid resistance of more than a few tenths of a percent is treated as a reject rather than a calibration adjustment; a low isolation reading points to a contaminated surface, a lifted grid, or a solder splash at the pad. Visually examine whether the strain gauge appears even before the leads are soldered — a visible wrinkle at this stage cannot be fixed by post-cure pressure [S5].
Gage selection parameters that drive the install sheet

The installation and operating characteristics of a strain gage are affected by the strain-sensitive alloy, the backing material (carrier), the grid resistance, the gage pattern, the self-temperature-compensation number, the gage length, and the options package — Vishay's TN-505 selection document lists these as the seven selectable parameters that have to be matched to the environment before the install sheet is written [S3][S4]. On a pressure sensor diaphragm a constantan-alloy foil with a polyimide carrier is the default configuration; a weldable gage on a steel structure uses a different carrier and a different self-temperature-compensation number matched to the coefficient of thermal expansion of the substrate. A mismatched compensation number is the most common reason a "good" install drifts off zero on a temperature cycle, and it cannot be corrected by shunt-cal after the fact.
Protection, coatings, and the strain-relief loop
Field installations are protected from moisture, oil, dirt, and light mechanical damage by a flexible coating system — the DTIC strain-gauge manual recommends two or more coats of a polyurethane or neoprene coating (Micro-Measurements M-Coat A or D, or an equivalent) for indoor use, with a butyl-rubber or epoxy overcoat added for harsher exposure [S6]. The cables leaving the gauge need a strain-relief loop: mechanically protecting the installation cables also protects the strain gauge from external impact and from lead-fatigue fracture at the terminal [S5]. Special strain-gauge terminals (clamped or soldered, never twisted) improve reliability at the lead-transition point, which is the most common field-failure location in published failure analyses. On a pressure transmitter head the lead transition is potted in a stress-relief boot before the housing is sealed — the field equivalent is a butyl loop cemented to the substrate 20–30 mm from the pad.
Subsea, pressure-cycled, and chemically aggressive environments

Pressure cycling causes micro-leaks and progressive ingress at the housing and conduit seals; compression of the housing itself can compromise the seal and produce a sudden data offset after pressurization, or a gradually worsening noise floor in the Wheatstone-bridge output. For subsea applications, purpose-built pressure-rated transducers (subsea load cells, underwater torque sensors) are preferred over exposed foil gauges, with pressure-compensated housings and qualified cable penetrations. The same logic applies to chemical-plant diaphragm seals where a flush-mount industrial valve body isolates the foil from the process — at that point the gauge is bonded to a metal diaphragm in a controlled factory environment, not in the field, and the install sheet is replaced by a factory calibration certificate. [S1]
Foil vs semiconductor vs optical fiber — which to install
Three sensing families dominate the install sheet. Foil gauges (HBM stress-analysis and transducer series, Vishay Micro-Measurements) are the default for room-temperature structural and pressure work, with the largest installed base and the most published install notes. Semiconductor gauges give a higher gauge factor but are temperature-sensitive and harder to bond; they show up where output voltage is at a premium and the environment is controlled, for example in servo motor torque sensors and miniature load cells. Optical-fiber strain sensors (Bragg-grating or distributed) carry the signal as light, are immune to EMI, and survive the high-temperature and high-voltage environments that kill foil — but they need an interrogator, fusion splices, and a different install skill set. Decision rule: pick foil for static structural and pressure work, semiconductor for high-output low-drift transducer bodies, and optical fiber where EMI, voltage, or temperature rules out the other two. Integration downstream: the four-arm Wheatstone-bridge output of a bonded foil gauge is amplified and linearized inside a pressure transmitter head, or fed directly into a PLC analog-input module for test-rig and structural-health-monitoring use; on a Coriolis flow meter the same bridge topology is bonded to the curved flow tube and resolves mass flow from the inertial force on the tube. [S2]
The Micro-Measurements, Vishay, and HBM technical-note corpus treats surface prep, lead transition, and seal as the three controlled nodes; the next reference for the practitioner is the M-Line accessory catalog (terminals, coatings, lead wire) and the shunt-calibration procedure for in-service bridge verification. For hostile-environment work, switch to a factory-bonded foil inside a sealed pressure-transmitter head, or to a fiber-Bragg sensor with an EMI-immune optical readout — and document the install sheet so the next maintenance crew does not have to re-derive what was bonded where.