01
Fibre burst
Netting theory, with fibre strength knocked down to 0.75 × measured after Tew (1995). The method is checked against two published burst-test series, Yurdakul (2022) and Karpuz (2005), and lands either on the safe side or within 15%.
Software · Motor hardware, from case tube to proof test
RustCase
Designs the hardware a solid motor is built in: the case tube — machined metal or filament-wound composite — the forward closure, the aft closure that carries the nozzle, what retains them, and their O-rings. It checks every part against the motor’s pressure and heat, gives the case a rated maximum operating pressure, writes the hydrostatic proof-test procedure, and exports the parts for a machine shop with a bill of materials. Engineering estimates for amateur and research motors, not a pressure-vessel code. Coming soon.
The RocketSalvo family
Design the propellant, shape the motor, build the hardware, measure the burn.
RustCore predicts the pressure a motor reaches. RustCase is what that pressure has to be contained by, so the two share the same numbers rather than meeting at a value copied across by hand.
RustPEP
Design the propellant
Predicts the recipe’s performance and holds its burn-rate law, r = a·Pⁿ.
About RustPEPRustCore
Shape the motor
Designs the grain and core, and predicts the pressure the case will have to hold.
About RustCoreRustCase
Build the hardware
Designs and checks the case, closures, retention and seals, and rates the assembly.
How a case is designedRustStand
Measure the burn
Records the static test, and which case the motor was fired in.
About RustStand
Why it exists
A case is not a tube with ends on it.
The parts set each other’s dimensions. A snap-ring groove removes wall that the burst check needs. The screws retaining a closure have to sit where there is metal to carry them. An O-ring needs a gap small enough not to extrude into and enough squeeze to seal. Check the parts one at a time and the assembly can still be wrong.
And heat is usually what decides it. Aluminium loses strength long before it melts, so a case that passes on pressure at room temperature can still fail late in a burn, or during the soak after it.
- Case tube29 mm to 12 in, machined metal or filament-wound
- Forward closureBulkhead bending checked
- Aft closureCarries the nozzle; lip bending checked
- RetentionScrews, snap rings or threads; pins or bonded in composite
- SealsO-rings, from the Parker handbook
How it works
From the motor’s pressure to parts on a drawing.
1 · Size it
Start from the pressure
Pick the motor size and tube stock, or type your own. RustCore supplies the pressure the motor will reach and when the burn exposes the liner, so the case is designed against the motor it will actually hold rather than a round number.
2 · Check it
Around forty checks, with margins
Tube burst and yield, screw shear, bearing, tear-out and net section, snap-ring ratings, thread shear, closure and nozzle-lip bending, and O-ring squeeze, fill, stretch and extrusion gap. Each one reports its margin, so you can see which part is closest to the edge.
3 · Heat it
Through the burn, and the soak after
Wall temperature as a transient through the liner, including the soak after burnout when thrust has stopped but the heat is still arriving. Strength loss comes from the MIL-HDBK-5J half-hour curves, and the liner is sized to keep the wall inside what the alloy can still hold.
4 · Rate it
A MEOP, then a proof test
A rated maximum operating pressure with the check that governed it and the conditions behind it, then the hydrostatic proof-test procedure — 1.5 × MEOP with water by default — and a record of the tests the case has passed.
5 · Build it
Parts a shop can work from
STEP with exact geometry, PDF drawings, DXF and STL, and a bill of materials as CSV and as a web page with supplier links. Part numbers appear only where they have been checked against the supplier’s listing.
In development · where it stands
Version 0.4.0 runs on the developer’s machine.
The desktop app works: it opens its own window and shows the Design page with the case’s rating and its checks, and the Design, Closures, Heat, Proof, Parts and BOM, Link and Help pages are all on screen. The same interface has been driven end to end in the test harness, all 61 tests pass, and FreeCAD opens every STEP part it exports.
Recent builds added a 3D view of the assembly that you can rotate and cut away, with RustCore’s grains and cores drawn inside the case; a liner that runs over spigots on the bulkhead and nozzle; a 304 stainless washer between a graphite nozzle and its snap ring, with its own drawing and bill-of-materials line; and PDF drawings in Rocket Salvo colours. Version 0.4.0 also fixed a hang and about thirty bugs found in a code review.
It has not been released or installed anywhere else. There is no released installer, so nothing described here is available to anyone but its developer yet.
Filament-wound composite tubes are on screen too, and are described in their own section below rather than folded into the metal case: they fail differently and are checked differently.
Filament-wound tubes
A composite case is not an aluminium one with better numbers.
Set the tube’s construction to filament-wound and RustCase designs a wound tube instead of a machined one: a fibre system of carbon/epoxy or E-glass/epoxy, the helical angle, the count and thickness of the ±α helical and 90° hoop layers, and the resin with its Tg — or it chooses the layup for you. A sealed liner, a rubber bladder or a metal one, is a tick box, because it changes which failure governs.
02
Weep, which usually governs
First-ply failure by laminate theory with Tsai-Wu. A thin wound wall cracks in the matrix and leaks at around 40% of its burst pressure, long before the fibres break. Without a sealed liner this is normally what sets the rating — so a wound case is often limited by the resin letting gas through, not by the fibre strength that made composite attractive.
03
Temperature, from the resin
The wall limit is the resin’s wet Tg less 28 °C, after MIL-HDBK-17. Matrix strength falls off following Chamis’s model and fibre strength loses up to 15% near Tg. Room-cure laminating epoxies such as West System sit near 55 °C Tg, which leaves a wall limit around 27 °C, and the app says they are unsuitable rather than letting you design with one.
04
Joints that do not cut fibres
Closures are retained by radial pins through the wall, or bonded in with EA 9394, EA 9430, DP460 or DP420. Snap rings and threads are refused for a wound tube, because the groove or thread would cut the fibres carrying the load. Pins are checked in bearing and net section, with at least 3 hole diameters to the end and 4 between holes; bonded joints in lap shear against temperature over a Hart-Smith effective length, with a factor of 2.
What it is designed to do
Planned for the first release.
01
The whole assembly, not just the tube
A case is a tube, a forward closure, an aft closure carrying the nozzle, and something holding the ends in — radial screws, snap rings or threads — with the O-rings that seal them. RustCase designs all of it together, because the parts set each other’s dimensions: the groove for a snap ring removes wall the burst check needs, and the screws retaining a closure have to sit where there is metal to carry them. Sizes run from 29 mm to 12 in, from presets or typed in, with tube stock from an OnlineMetals catalog in 6061-T6, 7075-T6, 7075-T73 or 7050-T7451.
02
Around forty checks, each with its margin
Tube burst and yield. Screw shear, bearing, tear-out and net section. Snap-ring ratings from the Rotor Clip HO and KG series. Thread shear. Bending in the bulkhead and at the nozzle lip. O-ring squeeze, fill, stretch and extrusion gap, worked from the Parker handbook. Every check reports its margin rather than a pass stamp, so you can see which part is closest to the edge and by how much. Design factors default to 2.0 on ultimate and 1.5 on yield and are yours to change. Screws can be sized automatically, in alloy steel or 18-8 stainless, imperial or metric.
03
Heat, which is what usually decides it
Aluminium loses strength long before it melts, so a case that passes on pressure at room temperature can still fail late in a burn. RustCase works the wall temperature through the liner as a transient, including the soak after burnout when the motor has stopped producing thrust but the heat is still arriving, and takes the strength loss against temperature from the half-hour curves in MIL-HDBK-5J. It sizes the liner to keep the wall inside what the alloy can still hold. Liner ablation and charring are not modelled, which means the estimate errs hot: it predicts a hotter wall than you will get, not a cooler one.
04
A rated MEOP, and the test that confirms it
The output is a rated maximum operating pressure together with the basis it came from — which check governed, at what temperature, under what factor — rather than a bare number. It then writes the hydrostatic proof-test procedure, by default 1.5 × MEOP with water, and keeps a record of the tests a case has passed. The procedure is the point: a calculation says a case of this design should hold, and a proof test is what establishes that this particular piece of metal, with its own machining and its own flaws, actually does.
05
Parts a machine shop can work from
STEP files with the exact geometry, PDF drawings, DXF for a waterjet or laser, and STL. The bill of materials comes as CSV and as a web page with links to McMaster-Carr, OnlineMetals, Rocket Motor Parts and Rotor Clip. Supplier part numbers appear only where they have been checked against that supplier’s own listing; most McMaster lines are search links rather than a specific part, so expect to confirm what you are ordering. Imperial or SI throughout.
06
Part of the RocketSalvo family
RustCore tells RustCase the pressure the motor will reach and when the burn exposes the liner; RustCase tells RustCore the case bore and the pressure rating it has to design inside; RustStand records which case a firing was fired in, so a case accumulates a history. RustCore 0.23 and later already do their part: they read RustCase’s cases and write case loads for it. RustCase itself is not released yet, so no one can use the exchange end to end today.
Specifications
Technical details
Planned; subject to change before release.
| Status | |
|---|---|
| Availability | Coming soon. In development; not yet available |
| Scope | Engineering estimates for amateur and research motors. Not a pressure-vessel code, and no substitute for a hydrostatic proof test |
| Price and licensing | To be determined |
| Release date | To be determined |
| What it designs (planned) | |
|---|---|
| Motor sizes | 29 mm to 12 in, from presets or typed in |
| Tube construction | Machined metal, or filament-wound composite |
| Tube alloys | 6061-T6 (default), 7075-T6, 7075-T73, 7050-T7451 |
| Tube stock | From an OnlineMetals catalog |
| Closures | A forward closure, and an aft closure carrying the nozzle |
| Retention | Radial screws, snap rings or threads in metal; radial pins or a bonded joint in composite |
| Seals | O-rings, worked from the Parker handbook |
| Units | Imperial or SI |
| Checks (planned) | |
|---|---|
| Tube | Burst and yield |
| Screws | Shear, bearing, tear-out and net section, with automatic sizing in alloy steel or 18-8 stainless, imperial or metric |
| Snap rings | Rotor Clip HO and KG series ratings |
| Threads | Thread shear |
| Closures | Bulkhead bending and nozzle-lip bending |
| O-rings | Squeeze, fill, stretch and extrusion gap |
| Design factors | 2.0 on ultimate and 1.5 on yield by default, editable |
| Reported as | A margin for each check, rather than a pass or a fail |
| Heat and rating (planned) | |
|---|---|
| Wall temperature | Transient through the liner, including the soak after burnout |
| Strength against temperature | MIL-HDBK-5J half-hour curves |
| Liner | Sized to hold the wall within what the alloy can still carry |
| Rating | A maximum operating pressure, with the governing check and its conditions |
| Proof test | 1.5 × MEOP with water by default, as a written procedure, with a record of the tests a case has passed |
| Pressures | Gauge, throughout the exchanged design file |
| Filament-wound composite tubes (planned) | |
|---|---|
| Fibre systems | Carbon/epoxy, wet-wound, or E-glass/epoxy |
| Layup | Helical angle, with ±α helical and 90° hoop layer counts and thickness, or chosen for you |
| Liner | Optional sealed liner, a rubber bladder or a metal one, which changes the governing failure |
| Fibre burst | Netting theory, fibre strength knocked down to 0.75 × measured, after Tew (1995) |
| Checked against | Two published burst-test series, Yurdakul (2022) and Karpuz (2005); lands on the safe side or within 15% |
| Weep | First-ply failure by laminate theory with Tsai-Wu, at roughly 40% of burst; without a sealed liner this usually sets the rating |
| Temperature | Wall limit is the resin’s wet Tg less 28 °C, after MIL-HDBK-17; matrix strength by Chamis, fibre strength losing up to 15% near Tg |
| Retention | Radial pins, or bonded with EA 9394, EA 9430, DP460 or DP420. Snap rings and threads are refused: they would cut the fibres |
| Pin joints | Bearing and fibre net section, with at least 3 hole diameters to the end and 4 between holes |
| Bonded joints | Lap shear against temperature over a Hart-Smith effective length, with a factor of 2 |
| Unsuitable resins | Room-cure laminating epoxies such as West System, Tg around 55 °C, leave a wall limit near 27 °C; the app refuses them for a motor case |
| Composite tubes: not modelled | |
|---|---|
| Pin-bearing strength | A derived estimate, not a published allowable — 360 MPa carbon, 200 MPa glass — and labelled UNVERIFIED in the app |
| Domes | Not modelled; the closures are metal plugs in a straight tube |
| Winding scatter | Not modelled. Hand wet winding varies: strength coefficient of variation as high as 17% |
| Also not modelled | Fibre slip and thickness build-up at the tube ends, impact damage, moisture, creep and bending |
| Exports (planned) | |
|---|---|
| Geometry | STEP with exact geometry, plus DXF and STL |
| Drawings | |
| Bill of materials | CSV, and a web page with supplier links |
| Suppliers linked | McMaster-Carr, OnlineMetals, Rocket Motor Parts, Rotor Clip |
| Part numbers | Only where checked against the supplier’s listing; most McMaster lines are search links |
| Not modelled | |
|---|---|
| Fatigue | Repeated firings of the same case are not assessed |
| Stress corrosion | Not assessed |
| Flight and buckling loads | Not assessed; the case is checked against the motor’s pressure and heat |
| Liner ablation and charring | Not modelled, so the heat estimate errs hot |
| Stress concentrations | Beyond the net-section checks, not assessed |
What a calculation cannot tell you
The proof test is not a formality.
A calculation says a case of this design should hold. It cannot know about the piece of metal in front of you — its machining, its tool marks, its own flaws. That is what a hydrostatic proof test establishes, which is why RustCase writes the procedure and keeps the record rather than treating the rating as the end of the job. Fatigue over repeated firings, stress corrosion, flight and buckling loads, liner ablation and stress concentrations beyond the net-section checks are all outside what it models.
Start with the motor that goes inside it.
RustCore designs the grain and predicts the pressure RustCase will have to hold, and is in beta today.