Tools and Helpers¶

This page covers the bench: hand tools, inspection, fixtures, instruments, software and the templates that save you rebuilding a spreadsheet somebody has already built.

A CubeSat lab does not need everything listed here on day one. The ordering that generally serves teams best is measurement before fabrication – a decent multimeter, a caliper and an oscilloscope earn their keep long before a reflow oven does – and buy the tool that makes a job repeatable rather than the one that makes it possible. Repeatability is what separates a spacecraft from a prototype.

Hand Tools and Assembly Equipment¶

Fastener tools¶

Calibrated torque driver. The single most important tool on this page. CubeSats live on M2.5 and M3 fasteners in aluminum, where the gap between "not tight enough" and "stripped" is small. A driver covering roughly 0.2–2 Nm handles most CubeSat work. Have it calibrated, and record the torque applied to every fastener. See Structure – Fasteners and Assembly.
Quality hex and Torx drivers. Cheap drivers round out socket heads, and a rounded fastener in a nearly finished spacecraft is a bad afternoon.
Thread repair kit – helicoils or thread inserts. Aluminum threads do not survive many cycles, and being able to repair one is much better than replacing a machined part.

Wiring and harness tools¶

Proper crimp tools, matched to the specific contact. A generic crimper produces joints that pass a continuity test and fail after vibration. This is the classic false economy in CubeSat assembly.
Wire strippers sized for the small gauges used in flight harness (typically AWG 24–30), which ordinary strippers nick.
Contact insertion and extraction tools for the connector families you use.
Heat gun for heat-shrink, and a supply of appropriately rated shrink and sleeving – check outgassing before anything flies. See AIT – Electrical Assembly and Harnessing.

Soldering and rework¶

Temperature-controlled soldering station with fine tips.
Hot air rework station for surface-mount work and component removal.
Flux, braid, and a good extraction fan. Fume extraction is a safety item, not a comfort one.
A reflow oven or hotplate if you assemble your own boards.

Solder alloy and tin whiskers¶

Choose the alloy deliberately. Pure tin finishes spontaneously grow single-crystal tin whiskers, and NASA's position after decades of investigation is blunt: the only certain way to avoid them is not to use pure tin plating, and alloying the tin with lead suppresses the growth. On the ground a whisker is a nuisance. In vacuum it is worse – a whisker short can strike a plasma capable of carrying hundreds of amperes, and a commercial satellite has been rendered non-operational that way.³

This is why spaceflight practice stayed with tin-lead while the rest of the industry moved to lead-free under RoHS, and why it matters to a CubeSat built from COTS parts. The aerospace exemptions in the legislation do not protect a board you did not specify: suppliers have switched to lead-free finishes without changing part numbers, and buyers with contracts prohibiting pure tin have received it anyway.³ Assume a COTS module has a lead-free finish unless its datasheet says otherwise. Conformal coating is a mitigation rather than a cure, and mixing lead-free parts with tin-lead solder has its own failure modes, so the decision is worth making before you buy rather than at assembly. There is a standard for this: NASA requires suppliers of printed circuit board assemblies to work to SAE GEIA-STD-0005-1A and GEIA-STD-0005-2A at control level 2C, the latter being the standard for mitigating tin whiskers in aerospace electronics.²

The space addendum to J-STD-001 names the alloys it expects: Sn63Pb37, Sn60Pb40, Sn62Pb36Ag2 and Sn96.3Ag3.7.¹ For hand work, Sn63Pb37 is the default – it is the tin-lead eutectic, melting and freezing at a single 183 °C with no plastic range, which is what makes a disturbed joint obvious rather than subtle. Sn62Pb36Ag2 is the one to reach for on silver-plated parts and thick-film terminations, where the silver already in the alloy slows the leaching that would otherwise dissolve the plating into the joint. Sn96.3Ag3.7 is the lead-free option the addendum allows, and it is not a drop-in: it melts around 40 °C higher and brings the whisker question back with it.

Solder paste and flux¶

Solder paste is powder suspended in flux, typically 88–91% metal by weight, and the powder is graded by particle size. IPC J-STD-005 sets the bands: Type 3 at 25–45 µm, Type 4 at 20–38 µm, Type 5 at 15–25 µm and Type 6 at 5–15 µm, so the smaller the number, the coarser the powder.¹

Pick by stencil aperture, not by ambition. The five-ball rule from IPC-7525 says the smallest aperture should span at least five of the largest particles, which puts Type 3 at roughly 225 µm minimum, Type 4 at 190 µm and Type 5 at 125 µm.¹ In component terms Type 3 covers down to 0402 imperial and handles most CubeSat boards; Type 4 is the usual choice at 0201 and for fine-pitch BGAs; Type 5 is for 01005 or for dispensing through a needle rather than printing. Use the coarsest type that releases reliably – finer powder has more surface area per unit mass, so it oxidizes faster, slumps more readily and has a shorter usable life. Paste lives in the fridge at 2–10 °C and needs to reach room temperature sealed, or it collects condensation.

Flux is classified by IPC J-STD-004 as four characters: base chemistry (RO rosin, RE resin, OR organic, IN inorganic), activity (L, M or H) and halide content (0 or 1). ROL0 is the familiar low-activity halide-free rosin. The space addendum narrows the field to rosin and resin bases at L0 or L1 activity, which rules out the water-soluble and organic-acid fluxes entirely.¹

No-clean is not no-consequence

No-clean flux is designed to leave a residue that is benign on a bare board. It is not necessarily benign under a conformal coating: rosin and resin no-clean residues can be incompatible with coatings and underfills, producing voiding or poor adhesion, and a coating applied over residue traps whatever is underneath it for the life of the mission. If the board is going to be coated – and on a CubeSat it usually should be – clean first and choose a flux you can actually remove. Isopropyl alcohol, ethyl alcohol and deionized water are the solvents NASA treats as standard, but aqueous cleaning is prohibited on silver-plated copper.²

Learning the skill¶

Hand soldering is a learned motor skill, and the fastest way to learn it is to watch someone competent do it slowly and then copy them. The written guides below cover through-hole and basic surface mount; the video channels are where rework lives, which is the part that matters once a board is populated and a part has to come off without lifting a pad.

What none of them cover is what makes a joint acceptable on flight hardware. That comes from a workmanship standard, not a tutorial: NASA cancelled its own soldering standards on 17 October 2011 and adopted the space addendum to IPC J-STD-001 instead, with NASA-STD-8739.6 covering implementation.⁴ A CubeSat flying as a secondary payload is rarely held to it, but the acceptance criteria are worth reading once – they are what a trained inspector is looking at, and they are more specific than any tutorial about wetting angles, fillet shape and what counts as a disturbed joint.

How to Solder: Through-Hole Soldering Link – SparkFun's tutorial, and the best written starting point: what solder is, what the tools do, the technique itself, and a gallery of joints that went wrong with the reason for each
SparkFun soldering tutorials Link – The rest of the series, covering surface mount, castellated holes, rework and wire work
Another Teaching Moment: Through-Hole Soldering Basics Link – Digi-Key's short written walkthrough, with a gallery of good and bad joints and a link to their demonstration video at the end. The most compact of the written guides
Adafruit Guide to Excellent Soldering Link – Tools, preparation, making the joint, surface mount, and a common-problems page that is the most useful part. Linked here as the single-page version
Mr SolderFix Link – Short, quiet, close-up demonstrations of rework: QFN and QFP removal, clearing filled holes with braid and liquid flux, lifting pins, repairing traces, replacing connectors. The channel to watch before attempting rework on a board you care about
iFixit's Soldering 101 Link – A single beginner video covering the whole workflow end to end, aimed at repair rather than assembly
SDG #066 Solder Through-hole Components Link – From SDG Electronics. Through-hole technique shot close enough to see what the solder is actually doing, which most beginner videos are not

ESD control¶

Non-negotiable for flight hardware, and cheap: an ESD mat, a wrist strap with a verified ground, ESD-safe tools and storage bags, and humidity control where the climate makes static a problem. The failures ESD causes are usually latent – the part works, then fails weeks later – which makes them nearly impossible to diagnose after the fact.

The numbers are worth knowing, because "ESD-safe" on a product listing means nothing on its own. NASA's implementation standard, which builds on ANSI/ESD S20.20, asks for a work surface between 10⁶ and 10⁹ Ω and the same range from surface to ground, a wristband between 800 kΩ and 1.2 MΩ, a wrist strap ground terminal under 1 Ω, and – the one most benches fail – a soldering iron tip to ground resistance under 5 Ω. Relative humidity sits between 30% and 70% for ordinary parts and 40% to 70% for the most sensitive. Wrist straps are checked before every use, not periodically.²

Two habits that standard would break. Pink poly antistatic bags are not acceptable as primary packaging for sensitive parts unless they are verified free of tertiary amines – the shielding bag is the one that counts, and pink poly is for physical protection. And an insulator within 300 mm of an exposed part needs a charge mitigation plan, which in a small lab usually means moving the coffee cup and the roll of tape rather than buying an ionizer.

Cleanliness and handling¶

Nitrile gloves, lint-free wipes, isopropyl alcohol, and covered storage for anything that flies. Gloves for everything, without the exceptions that gradually erode over a long build. See Structure – Cleanliness.

Inspection and Measurement¶

Digital calipers (0.01 mm resolution) as the everyday dimensional tool, and a micrometer where the tolerance warrants it. Rail dimensions are held to ±0.1 mm, which calipers can just about resolve and a micrometer resolves comfortably.⁵ See Structure – Tolerancing and Stack-Up.
A precision scale. Mass is a tracked budget item, and per-component measured mass is far more useful than datasheet values. Something reading to 0.1 g covers most CubeSat components; a 0.01 g scale is better for small parts.
Stereo microscope or digital inspection camera for solder joint inspection, connector examination and general "what actually happened here" work. A USB microscope is inexpensive and transforms board debugging.
Feeler gauges and pin gauges for clearance and hole checks.
A surface plate and height gauge, if you want to verify rail flatness and envelope dimensions without CMM access.
CMM access, if you have it, for the definitive dimensional verification – but a fit check in a test pod is the answer that actually matters. See Structure – Tolerancing and Stack-Up.
Mass properties measurement. Center of mass can be determined adequately with a knife-edge balance or a three-point scale setup and some arithmetic; commercial mass properties benches exist but are rarely justified at CubeSat scale. See Structure – Mass Properties.

Visual inspection remains the highest-yield inspection method for a small team. Photograph everything at every stage – it costs nothing and repeatedly turns out to be the only record of how something was before it was disturbed.

Jigs, Fixtures, and Test Aids¶

Fixtures are where a little design effort produces disproportionate returns, because they convert a delicate operation into a repeatable one.

Assembly and alignment jigs hold the stack square while standoffs are torqued, or locate a payload precisely during bonding.
Handling fixtures protect the rails – the most tolerance-critical and most easily damaged surfaces on the spacecraft – and give somewhere safe to set the satellite down.
Test fixtures and breakout boards exposing every bus and rail on accessible headers. Build these properly; you will use them for years. See AIT – Flatsat.
Vibration test adapters interfacing the spacecraft or its test pod to the shaker table.
Deployment test rigs, including gravity offloading where the geometry allows, since a mechanism that deploys on a bench may be relying on gravity to help.
Thermal test fixtures with low conductivity, so the fixture does not become the dominant heat leak in a TVAC run – a common source of misleading thermal balance results.

3D printing versus machining: printed fixtures are cheap, fast to iterate and perfectly adequate for handling aids and soft jigs. Machined fixtures are worth it where dimensional accuracy or stiffness matters, or where the fixture will see load. Print first, machine what proves worth keeping.

Design fixtures with the same care as flight hardware in one respect: they should make the wrong assembly impossible, not merely make the right one convenient.

Electrical and RF Test Equipment¶

Power¶

A bench power supply with adjustable current limit is the most important electrical instrument you own. The current limit is what prevents a wiring error from destroying a board.
A solar array simulator is what a bench supply is not. A supply provides effectively unlimited current at a fixed voltage; a solar array does not, and that difference is precisely what MPPT and brownout behavior depend on. A programmable supply with an I-V curve capability, or a purpose-built simulator, changes the fidelity of EPS testing completely. See EPS – Ground testing.
An electronic load for characterizing battery packs and converters.
Current probes or shunt-based measurement for per-channel current, ideally logged over hours so that mode transitions and duty cycles are visible.

Signals¶

Oscilloscope. Two channels is a minimum, four is much better – power rail, clock and two bus signals simultaneously is a common need. Bandwidth requirements for CubeSat work are modest; channel count and memory depth matter more.
Logic analyser, ideally with protocol decoding for I²C, SPI, UART and CAN. A low-cost USB logic analyser is one of the highest-value-per-euro instruments available for debugging exactly the bus problems described in OBC – Interfaces and Buses.
Multimeter, or several. Continuity and isolation checks on harness are performed constantly.

RF¶

Spectrum analyser for verifying transmitter output, spurious emissions and, importantly, the RF silence period during inhibit verification. A low-cost analyser or even an SDR with calibrated attenuation covers a lot of ground.
Vector network analyser for antenna matching and filter characterization. Inexpensive VNAs covering VHF/UHF are now widely available and are entirely adequate for CubeSat antenna work.
Attenuators and dummy loads, so that end-to-end RF testing can be done without transmitting into the air – which also keeps you legal. See AIT – Mission Simulation.
An SDR doubles as a receiver, a spectrum monitor and a signal-analysis tool. See Ground Segment.

Practical tradeoffs for small teams: buy the current-limited supply, the logic analyser and the caliper first. Borrow the shaker table and the TVAC chamber. Second-hand test equipment is generally excellent value, since instruments age well. And an instrument you cannot use is worth less than a cheaper one you understand.

Software Tools¶

Mechanical CAD and analysis¶

FreeCAD Link – Parametric CAD with an integrated FEM workbench; increasingly capable and free for student teams. Open source (LGPL)
CalculiX Link – Three-dimensional structural finite element solver with static, dynamic and thermal capability. Open source (GPL)
PrePoMax Link – Pre- and post-processor giving CalculiX a modern user interface. Open source
Code_Aster Link – Structural mechanics and thermomechanics solver developed by EDF R&D. Open source (GPL)
Elmer FEM Link – Multiphysics finite element software. Open source (GPL)

Commercial options – SolidWorks, Fusion, Onshape, Ansys, Nastran/Femap – dominate in industry and most offer free or heavily discounted educational licenses. See Structure – Toolchains.

Electronics design¶

KiCad Link – Free, open-source PCB design suite with schematic capture, PCB layout, 3D viewer, integrated SPICE simulation and a Gerber viewer; the default choice for open-hardware CubeSat projects. Open source
Build a CubeSat hardware Link – Fully open KiCad designs and manufacturing files for every node in the Build a CubeSat stack. Open source (CERN OHL-S 2.0), in development

Using KiCad has a specific advantage for this community beyond cost: open-hardware CubeSat designs are overwhelmingly published as KiCad projects, so it is the format in which you can actually read other people's work. See the open-source missions in CubeSat Missions.

Orbital mechanics and mission analysis¶

NASA GMAT (General Mission Analysis Tool) Link – Mission design, optimization and navigation tool used for flight mission operations as well as analysis and teaching. This is the current R2026 catalog entry; older release entries are still indexed and easy to land on by mistake. Open source
Basilisk Link – Modular astrodynamics simulation framework from the University of Colorado AVS Lab, well suited to spacecraft dynamics and ADCS simulation. Open source
CelesTrak Link – Orbital element sets, SGP4 references and a large body of astrodynamics documentation. Free

Python has become the default language for this work, with mature libraries for SGP4 propagation, coordinate transformations and pass prediction. For a CubeSat team, a short Python script that propagates a TLE with SGP4 and predicts passes is usually more useful than a large commercial package, and much easier to integrate with the ground segment.

Thermal analysis¶

See Thermal – Tools for open-source and commercial thermal modeling options, including SATMO and the single-node approach.

Firmware and software development¶

Toolchains: GCC for ARM, PlatformIO, vendor IDEs, and the build systems around them. Whatever you choose, make the build reproducible – a build that only works on one person's laptop is a bus-factor risk. See Flight Software – Documentation and Maintainability.
Debuggers: a JTAG/SWD probe is essential. Plan physical access to the debug header early. See OBC – Integration and Testing.
Version control: Git, for everything – firmware, ground scripts, documents, test procedures, CAD where practical.
Simulation and test harnesses: the ability to compile flight software for the host and run it against simulated hardware is worth building deliberately.

Calculators and Reference Tools¶

A small but growing collection of templates and worked examples that save you from rebuilding the wheel. Most are free; a few are reference documents rather than fillable templates, but all are useful starting points.

Stack builders¶

cubestack.dev (currently in beta) (Patrik Senkyr) – Interactive, browser-based PC/104 stack configuration and validation tool. Includes a board library, a physical stack editor and power budget simulator.

Power Budget Templates¶

A power budget tracks energy consumption and generation across mission modes. See EPS – Power Requirements and Budgets for the underlying methodology.

Artemis CubeSat Kit Power Budget (Hawaii Space Flight Lab) – multi-sheet template covering component-level draws, operational modes, and generation/storage balance. Free to copy in Google Sheets or download as .xlsx. The companion textbook chapter at pressbooks-dev.oer.hawaii.edu/epet302 walks through how to use it step by step.
BIRDS Project power analysis (Hari Ram Shrestha, LaSEINE, Kyushu Institute of Technology, March 2022) – published power budget for a 1U mission, working through beta angle and eclipse duration explicitly: 36 minutes of eclipse at β = 0°, 35 minutes at β = 30°, none at β = 73°. It also carries measured generation data from the flown Tsuru satellite across three beta angles, which is what makes it worth reading – less a template than a worked example of what a defensible analysis looks like, checked against orbit.
"Power Budgets for Mission Success" (Craig Clark & Ritchie Logan, Clyde Space, 2011) – slide deck on the Cal Poly CubeSat workshop archive. Practical walkthrough on estimating orbit average power, managing loads, and avoiding negative power budgets.

Mass Budget Template¶

CubeSat Resources Mass Budget Template – annotated spreadsheet with component-level mass tracking and form-factor limits for 1U through 12U, checked against the CDS 14.1 figure of 2 kg per U. Margin is applied per component according to how well the mass is known – 30% where the figure comes from a datasheet, down to 5% for a measured flight part – so the budget tightens as evidence accumulates rather than by decree. See Systems Engineering – Margin philosophy.

RF Link Budget Template¶

Jan King Link Budget Calculators (Jan King, VK4GEY/W3GEY / AMSAT-UK) – a collection of Excel spreadsheets for satellite link budget analysis, freely available for amateur and non-commercial use. The flagship tool is the AMSAT/IARU Annotated Link Model System, a detailed multi-sheet workbook that walks through every gain and loss term – path loss, antenna gains, noise figure, Eb/N0, modulation – with explanatory notes alongside each calculation. Widely cited in CubeSat comms literature and used as a reference in the Hawaii CubeSat textbook. Free to download; .xls format. See also Comms – Link Budget and the link budget glossary entry.

Other Budgets¶

Thermal: see Thermal – Tools for the single-node Python script and SATMO, both of which serve as worked thermal estimation examples rather than blank templates.
Data budget: data generated per orbit versus downlink capacity. There is no widely used template for this – a spreadsheet with payload data rates, duty cycles, pass count, pass duration and link rate covers it, and it is worth building early because it constrains the mission more than teams expect. See Comms – Expected Data Rates.
PV budget: a panel-by-panel photovoltaic generation analysis over the orbit and over the mission. See EPS – Estimating generation.
Pointing budget: sensor noise, estimation error, control error, mounting misalignment and thermal distortion, combined and compared against the requirement. No template exists that is worth recommending; a spreadsheet with one row per error term is the usual approach. See Systems Engineering – The main budgets.
Battery sizing: the relation in EPS – Sizing is straightforward enough to implement in a few spreadsheet cells; the difficulty is in the assumptions rather than the arithmetic.

Jan King Link Budget Calculators Link – Collection of Excel link budget spreadsheets including the AMSAT/IARU Annotated Link Model System

Automation and Workflow Helpers¶

Automation is where a small team buys back time, and the returns compound over a multi-year project.

Automated functional tests. A scripted test that runs in twenty minutes will be run after every change; a manual one that takes a day will be skipped under schedule pressure. This is the highest-return automation available to a CubeSat team. See AIT – Functional and Integration Testing.
Continuous integration for firmware and ground software – build on every commit, run the host-based unit tests, flag regressions. Free for open repositories.
Documentation CI. This site is itself an example: MkDocs building from a Git repository on every push. The same pattern works for mission documentation, and it means the published document always matches the repository.
Test data collection scripts. Instrument the bench: log currents, temperatures and telemetry to a file automatically rather than by hand. Long-duration tests are only useful if the data is captured.
Reproducible environments. Containerized or scripted toolchain setup, so a new team member is productive in an hour and a build from three years ago still works. On a project with complete team turnover every few years, this is a reliability measure.
Automated ground operations – pass scheduling, tracking, decoding and ingest. See Ground Segment – Automation and Operations.

The general principle: anything done more than about five times should be automated, and anything whose manual execution could damage flight hardware should be automated sooner than that.

Documentation and Knowledge Management¶

University CubeSat teams turn over on the same timescale as the project itself – see Systems Engineering – Organizational pitfalls – so knowledge management is not administrative overhead. It is the mechanism by which the project survives its own team.

Lab notebooks and build logs. Dated, specific, and including the things that did not work. "Tried X, it failed because Y" is often more valuable than the record of what eventually worked.
Written procedures for every repeated operation – assembly steps, test setups, operations. Two people: one performing, one recording.
Checklists for anything with irreversible consequences. Pre-integration, pre-delivery, pre-pass. Checklists are how aviation stopped losing aircraft to forgotten steps, and the reasoning transfers directly.
Version control for hardware too. Board revisions, mechanical drawings, BOMs and assembly instructions all belong in the repository alongside the code. See Systems Engineering – Configuration and Change Management.
A single source of truth for interfaces. The command and telemetry dictionary, the ICD set, and the pin assignments should live in one place that both flight and ground software derive from.
An anomaly log capturing every unexpected behavior with its investigation and resolution. See Systems Engineering – Lessons Learned.
Onboarding documentation so a new member can get productive without a founder's attention. Write it the first time somebody joins, then keep it current.

A useful test: if the three people who know most about the project left tomorrow, could the remaining team finish it? For most CubeSat projects the honest answer is no, and closing that gap is documentation work.

👉 Please consider contributing!

Powder size bands are from IPC J-STD-005A Table 3-2, as reproduced in Karl Seelig and Tim O'Neill, Size Matters: The Effects of Solder Powder Size on Solder Paste Performance, SMTA. Free PDF. The same paper sources the five-ball rule from IPC-7525 and the minimum aperture sizes derived from it, the component-size guidance, and IPC-7525's recommended area ratio above 0.66. The flux designator scheme is IPC J-STD-004; the space addendum's restriction to rosin and resin bases at L0 or L1 activity and its named alloys (Sn60Pb40, Sn62Pb36Ag2, Sn63Pb37, Sn96.3Ag3.7) are reported by Sierra Circuits, free, a vendor summary rather than the standard itself – J-STD-001JS is paywalled, so confirm against it before writing either into a procurement specification. The conformal coating incompatibility of rosin and resin no-clean residues is described by Indium Corporation, free, also a vendor source. ↩↩↩↩
NASA Office of Safety and Mission Assurance, NASA-STD-8739.6B, Implementation Requirements for NASA Approved Workmanship Standards, approved 4 February 2021. Free PDF. Section 6.3.2 requires suppliers of printed circuit board assemblies to conform to SAE GEIA-STD-0005-1A and GEIA-STD-0005-2A at control level 2C for lead-free materials; GEIA-STD-0005-2A is titled Standard for Mitigating the Effects of Tin Whiskers in Aerospace and High Performance Electronic Systems. Section 7.2.5 gives the ESD values quoted here: work surface and surface-to-ground resistance of 10⁶ to <10⁹ Ω, wristband 800 kΩ to 1.2 MΩ, wrist strap ground terminal <1 Ω, soldering iron tip to ground <5 Ω, and relative humidity of 30-70% for HBM Class 1A and 40-70% for Class 0A. Section 7.2.5.3.2 requires a functional wrist strap check before each use; 7.2.4.3.4 restricts topical antistatic packaging such as pink poly; 7.2.4.2.2 requires a charge mitigation process for essential insulators within 12 inches of a sensitive item. These are NASA requirements for mission hardware, not a floor a CubeSat has to meet, but they are the numbers behind the practice. ↩↩↩
NASA Electronic Parts and Packaging Program, NASA Tin Whisker (and Other Metal Whisker) Homepage. Free. The standing NASA reference collection on the subject. Jay Brusse, Tin Whiskers: Revisiting an Old Problem, NASA GSFC. Free PDF. States that the only sure way of avoiding tin whiskers is not to use pure tin plating, that alloying tin with a second metal reduces whisker propensity, and that US military specifications introduced prohibitions on pure tin plating in 1992–1993 after a series of whisker-related failures. NASA GSFC, The Continuing Dangers of Tin Whiskers. Free PDF. Source for the vacuum plasma arc capable of carrying hundreds of amperes and for a commercial satellite rendered non-operational by a whisker short. NASA NEPP, Lead-Free Solder Body of Knowledge Report, 2005. Free PDF. Source for aerospace exemptions not protecting assemblies in practice, suppliers switching to lead-free without changing part identification, and buyers receiving pure tin finishes despite contractual prohibitions. Dated, but the procurement problem it describes has not gone away. ↩↩
NASA Office of the Chief Engineer, NASA-STD-8739.3, Soldered Electrical Connections. Free. The standard itself is cleared for public access and remains a readable statement of hand-soldering acceptance criteria, but it is cancelled: NASA adopted the space applications addendum to IPC J-STD-001 on 17 October 2011 and published NASA-STD-8739.6 to cover implementation across programs. Its scope is hand and wave soldering of through-hole assemblies and discrete wire and cable terminations; surface mount was the separate NASA-STD-8739.2, cancelled on the same date. J-STD-001 itself is an IPC standard and is paywalled. ↩
Exolaunch, EXOpod Nova User Manual, Rev. 1.2 (June 2024). Free PDF. Openly published deployer manual covering 1U–16U, with mass and center-of-mass allowances, rail dimensions and protrusion limits – it quotes ±0.1 mm on rail width and ±0.5 mm on rail length. Section 1.2 states that where the manual conflicts with the CDS, the manual takes priority. ↩