Skip to content
AIRSHIP.DIYSource

What does it cost, how long does it take, and where would you do it?

Could you actually build it?

A bill of materials priced at what an individual is actually charged, a labour estimate cross-checked two ways, and the building. This is the page where the answer is no, and the reason is not the airship.


The answer on this page is no

Every other page asks whether the vehicle works. This one asks whether it can exist, and it is the only chapter that ends in a refusal. The refusal is not about the airship.

Not buildable by one or two people as drawn

NOT BUILDABLE BY ONE OR TWO PEOPLE AS DRAWN. The physics closes: 3.3 million dollars of materials makes a vehicle that flies for a year and floats. The build does not close, and it fails on the building rather than on the airship. A 133 m shed costs 7.4 times what goes inside it, cannot be rented, and is required because a rigid airship cannot be assembled in weather. That is a well-supported no, and it points at exactly two things worth studying: a build sequence that never leaves a bare hull outdoors, and a vehicle small enough to fit a building that already exists.

Materials
$3.33M
$2.01M to $5.79M
The building
$24.50M
7.4x everything that goes inside it
Labour
47.8 kh
12 years for two people, full time
Held broadside by two
0.90m/s
1.7 knots. Not a wind, a draught.

CAPITAL

$3.3M of materials and $24.5M for the building is $27.8M before any labour is paid for, against about $1M an individual can plausibly raise. The building is the larger half.

SCHEDULE

48 thousand hours is 12 years for two people full time, and full time is not what two people building in their own lives do. At the ceiling of 10 years this needs 3 people, which is no longer a two-person project.

THE BUILDING

133 by 47 by 42 m clear, and you cannot rent one. The airship sheds that remain are museums, film studios or in use, and the last purpose-built one cost EUR 78M. This, not the physics, is what has stopped every individual since 1930.

What it is made of

Priced at what an individual is actually charged, in August 2026. Wholesale is not available to one person, and quoting it would flatter the largest line by a factor of seven.

  • Gas cell barrier laminate$698k15,502 m2
    15,502 m2 at 45 USD/m2. THE LARGEST LINE, AND THE ONE WITH NO PUBLISHED PRICE. Priced off a Dyneema composite sailcloth matched on areal mass, which is a proxy for manufacturing difficulty and not for function. The low and high span a factor of six because that is the honest state of knowledge.$310k to $1.86M
  • Carbon fabric for the frame$342k2,952 kg
    2,952 kg at 116 USD/kg. Retail. The same fibre as commodity tow is $35/kg, a factor of 3.3 below, and no individual is offered it. Substituting bought pultruded tube replaces this line, the epoxy line and the consumables line at once.$222k to $461k
  • Outer cover fabric$340k8,508 m2
    8,508 m2 at 40 USD/m2. The only published airship-specific fabric price found anywhere, and it comes from a builder rather than a mill. Also the rain catchment surface, so it is the cheapest square metre on the vehicle in terms of what it returns.$204k to $476k
  • Fuel cell stack and balance of plant$159k30 kW
    30 kW at 5,300 USD/kW. Retail small-stack pricing. Automotive stacks at volume are two orders of magnitude cheaper per kilowatt and are not sold to individuals.$127k to $191k
  • Laminating epoxy$138k2,308 kg
    2,308 kg at 60 USD/kg. US retail and EU hobby pricing on the identical resin system differ by more than two to one, and neither is a quotation for three tonnes.$92k to $208k
  • Flexible photovoltaic modules$138k139,456 W
    139,456 W at 0.99 USD/W. 964 m2 at 144.6 W/m2. Cutting the array coverage to what the mission needs took this line down with it, which is the one place in the bill where a physics correction saved money.$104k to $173k
  • Fin structure and control surfaces$127k1,102 kg
    1,102 kg at 116 USD/kg. Fibre only, at the same retail price as the frame. The fins are large because the Munk moment is destabilising at every angle of attack.$83k to $172k
  • PEM electrolyzer$120k40 kW
    40 kW at 3,004 USD/kW. IEA installed capex with a 35 percent small-scale premium added explicitly rather than folded in.$90k to $150k
  • Gondola, keel and compartment structure$111k963 kg
    963 kg at 116 USD/kg. Fibre only, for the 1350 kg of load-bearing shell the arrangement carries in the gondola and keel. It is not part of the frame line, and leaving it out was leaving out the part of the vehicle a person actually stands in.$72k to $150k
  • Vacuum bagging consumables$38k2,404 m2
    2,404 m2 at 16 USD/m2. Single use. Every square metre bagged is film, peel ply and breather thrown away, and this line disappears entirely if the frame members are bought rather than laid up.$27k to $50k
  • Engine and generator set$36k1 set
    1 set at 35,600 USD. The third leg of the powertrain, and the one the failure analysis leans on when the array and the fuel cell are both unavailable. A certified light-aircraft engine plus a generator of the matching rating.$28k to $43k
  • Lithium iron phosphate storage$35k150 kWh
    150 kWh at 230 USD/kWh. Retail. The global average pack price is less than half this and it is a price no individual is offered.$28k to $41k
  • Propulsors: motors, controllers, nacelles and vectoring$32k72 kW
    72 kW at 450 USD/kW. 4 units. Built up from retail listings rather than a quotation, and the vectoring gimbal is the part nobody sells off a shelf, which is why the band is wide.$18k to $58k
  • Hydrogen, first fill$32k2,484 kg
    2,484 kg at 13 USD/kg. 2484 kg fills 29130 m3 at the sea level fill fraction. Helium in the same volume would cost 25 times as much for 8 percent less lift, which is the cost half of an argument the safety chapter makes on other grounds entirely.$10k to $62k
  • Propellers$16k4 off
    4 off at 4,071 USD each. 6.0 m diameter. Priced off a two-blade carbon propeller for a light aircraft, which is a fraction of the size, so this is a floor rather than an estimate.$10k to $23k
  • Hydrogen storage vessels$12k17 kg stored
    17 kg stored at 700 USD/kg. Sized to hold 17 kg, which is the electrolyzer's output over the longest run of nights the energy balance has to bridge. The DOE cost record is a projected high-volume figure; an individual buying a handful of Type IV cylinders pays more, and how much more is not published.$7k to $21k
The solid bar is the nominal, the faint band behind it is the low to high range. Open a row for what it is and where the number comes from. Quantities come from the same mass statement that sizes the ship, so the bill cannot describe a different vehicle from the one in the cutaway.

It is mostly surface, and surface is sold by the metre

The intuition is that the powertrain is the expensive part. It is not. The three largest lines are gas cell barrier laminate, carbon fabric for the frame, outer cover fabric, together 58% of the named subtotal, while the fuel cell, the electrolyzer and the battery come to less than a fifth between them.

The whole vehicle is $0k per kilogram of gross weight. A business jet is about $1,500/kg and a cruising yacht about $50/kg, and an aerospace structure built out of retail materials belongs exactly where this lands.

The largest line has no published price

Nobody lists a price for 15,000 m² of 0.21 kg/m² para-aramid and metallised PET airship cell laminate, because nobody has bought any this century. It is priced off a Dyneema composite sailcloth matched on areal mass, which is a proxy for how hard it is to make and not for what it does.

The range on that one line spans a factor of six, which is wider than the gap between first place and fourth. A quotation from a barrier film converter is the single most valuable phone call in the whole bill of materials.

LineQuantityUnit priceNominalLowHigh
Gas cell barrier laminate15,502 m245 USD/m2$698k$310k$1.86M
Carbon fabric for the frame2,952 kg116 USD/kg$342k$222k$461k
Outer cover fabric8,508 m240 USD/m2$340k$204k$476k
Fuel cell stack and balance of plant30 kW5,300 USD/kW$159k$127k$191k
Laminating epoxy2,308 kg60 USD/kg$138k$92k$208k
Flexible photovoltaic modules139,456 W0.99 USD/W$138k$104k$173k
Fin structure and control surfaces1,102 kg116 USD/kg$127k$83k$172k
PEM electrolyzer40 kW3,004 USD/kW$120k$90k$150k
Gondola, keel and compartment structure963 kg116 USD/kg$111k$72k$150k
Vacuum bagging consumables2,404 m216 USD/m2$38k$27k$50k
Engine and generator set1 set35,600 USD$36k$28k$43k
Lithium iron phosphate storage150 kWh230 USD/kWh$35k$28k$41k
Propulsors: motors, controllers, nacelles and vectoring72 kW450 USD/kW$32k$18k$58k
Hydrogen, first fill2,484 kg13 USD/kg$32k$10k$62k
Propellers4 off4,071 USD each$16k$10k$23k
Hydrogen storage vessels17 kg stored700 USD/kg$12k$7k$21k
Named lines total $2.38M. A further $950k is allowed for everything not itemised: fasteners, wire, fittings, valves, plumbing, wiring, instruments, avionics, tooling and the several hundred things a build discovers.

How long it takes

Two independent estimates: task by task from the areas and joint counts, and hours per kilogram of empty weight from composite homebuilt aircraft, which is the only body of experience there is for a small number of people building an aircraft structure out of carbon in a shed.

  • Systems, fitout and rigging15.9 kh
    8481 kg of gondola, keel, machinery, wiring, plumbing, controls and accommodation, at half the structural rate per kilogram because much of it is bought as assemblies. THE COVER, THE GAS CELLS AND THE PHOTOVOLTAIC MODULES ARE EXCLUDED: the first two have their own tasks above and would otherwise be charged twice, and the third is bought and bonded rather than built, so charging it a structural rate per kilogram would bill nearly five thousand hours for sticking down a thousand square metres of laminate.10.6 kh to 21.2 kh
  • Laminate the frame members12.2 kh
    12203 m2 of PLY placement, which is 6 plies over 1923 m2 of part. Estimating this on part area rather than ply area understates it by that factor, which is the commonest way a composite build schedule goes wrong. THIS LINE DISAPPEARS IF THE MEMBERS ARE BOUGHT.6.1 kh to 24.4 kh
  • Pattern, weld and test the gas cells4.7 kh
    13480 m2 of film across 12 cells, plus about 11.2 km of seam. EVERY METRE OF IT HAS TO HOLD, and how tightly is an open question rather than a number this model can state: the purity budget is set by nitrogen coming IN through the film, while a cell held a few hundred pascals above ambient loses hydrogen OUT through a defect and purges inward leakage as it does so. The two do not share an arithmetic, and the seam requirement is the leak rate one, which needs the sniff test to answer.2.7 kh to 8.1 kh
  • Assemble and joint the frame4.4 kh
    44 thousand lattice joints over 2778 m of girder run: 16 longitudinals and 16 frames, at a 250 mm lattice pitch. At six minutes each this is the single largest task in the build and it does not go away by buying anything.2.7 kh to 8.9 kh
  • Moulds, jigs and assembly fixtures4.0 kh
    YOU CANNOT LAY UP A FRAME WITHOUT THE THINGS THAT SHAPE IT. A mould for every distinct member section, a jig for every ring diameter, and a fixture long enough to hold a longitudinal straight while it cures. One-off composite work spends roughly 33 percent of its part hours on tooling, and this task was simply missing until the per-kilogram cross check refused to agree.2.0 kh to 8.1 kh
  • Inflation, ground trials and flight test4.0 kh
    Purge, inflate over about three days, weigh off, mast trials, tethered hops, and then the 40 hours of Phase I an amateur-built aircraft must fly solo before it may carry anyone. Zeppelin NT needed 800 flight hours over three and a half years to reach a type certificate, so this line is a floor and not a plan.2.0 kh to 10.0 kh
  • Pattern, seam and lace the outer cover2.5 kh
    7090 m2, done at height on staging over the whole hull.1.4 kh to 4.3 kh
The solid bar is the nominal, the faint band behind it is the low to high range. Open a row for what it is and where the number comes from.
Task by task
47.8 kh
27.5 kh to 84.9 kh
Per kilogram cross check
35.8 kh
Ratio 1.33. Two routes, one number.
Two people, full time
11.9years
Full time is not what two people building in their own lives do

Labour is per ply placed, not per part made

A 1.5 mm frame wall is six plies of the fabric an individual can buy. Estimating the layup on the surface area of the finished part rather than on the area of ply actually cut, wetted, placed and debulked understates it by that factor, and it is the commonest way a composite build schedule goes wrong.

The frame assembly line is worse and does not go away by buying anything: roughly 44 thousand lattice joints, each fitted, jigged, bonded and inspected.

Calibrate against the only comparable programme

Pathfinder 1 is 124 m against this hull’s 118 m. LTA Research began work at Moffett Field in 2017, in a hangar that already existed, with a professional workforce and funding from a Google founder, and first flew untethered on 24 October 2024. Seven years.

Any two-person schedule that comes out shorter than that is wrong, and this one does not: 12 years at hours nobody actually works, and roughly double that in evenings and weekends.

Where you would do it

A rigid airship is assembled indoors and cannot be assembled anywhere else. The frame is a lattice with no skin on it for most of the build and will not survive weather, and the finished hull is a sail that two people cannot hold in a breeze.

1.75 m133 m clear, 47 m span118 m ship, 35 m over the fins42 m
Section through the shed, ship inside, drawn from the same hull radii as the 3D view and the same clear dimensions the facility model computes. The person at the door is 1.75 m. The clear height is set by the fin tip and not by the hull crown, which is a 12 m difference and the sort of thing discovered on the day the doors will not clear.
Clear internal
133×47×42m
6,294 m² of floor
Steel hangar
$24.50M
Escalated from the only costed airship hangar in the literature
Complete base
$32.34M
Hangar, mast, tractor, two mules, ballast and mooring circle
Air supported
$6.43M
A quarter the price, and the blower must never stop

The airship is the cheap part

$3.33M of materials goes inside a $24.50M building, a factor of 7.4. The long wall carries 8.1 MN at the 1926 Air Ministry design wind, and that load rather than the span is why airship sheds cost like cathedrals.

You also cannot rent one. About six buildings of this size exist and every one of them is a museum, a film studio, or in use. The last purpose-built airship hangar cost EUR 78 million and its owner went bankrupt before the ship flew.

And 12 hectares outside it

A moored airship weathervanes around its mast, so the entire riding-out circle must be clear and level: 12 hectares at the radius Akron and Macon used at Lakehurst. The 1926 Air Ministry standard asked for 800 yards square for a mast with no shed at all.

This is what the water landing requirement buys back. A ship that never comes ashore needs no circle, no mules and no ground crew, and it weathervanes off a bow drogue by itself.

And then you have to hold on to it

The most counterintuitive result in the build chapter. This vehicle is safe in a gale and helpless in a breeze, and every ground operation has to be designed around that inversion.

Side area
2,705m²
Hull profile plus the vertical fins
Two people, broadside
0.90m/s
1.7 knots
Two people, bow on
5.2m/s
6x better, and it is the whole case for a mast
Crew to hold it at the Navy limit
99people
The Navy used 18, with a mobile mast and two mechanical mules
  • TWO PEOPLE CAN HOLD THIS SHIP BROADSIDE IN 0.90 m/s OF WIND, which is 1.7 knots and is not a wind, it is a draught. The 2705 m2 of side area is the whole problem.
  • Bow on it is 5.2 m/s, a factor of 6 better, which is the entire argument for single-point mooring: the ship must be free to weathervane at all times and must never be held across the wind.
  • At the 6.3 m/s the US Navy would still dock in, holding it broadside by hand needs 99 people. The Navy did it with 18 because they had a mobile mast and two mechanical mules; before that machinery existed, LZ-8 took 300 and was destroyed against the shed doors anyway.
  • BUT THE SHIP CAN HOLD ITSELF. Its own vectored thrust holds it bow-on in 17 m/s at zero airspeed, which is above the 6.3 m/s the US Navy would dock in. It does not help broadside, where it manages 2.6 m/s, and no plausible installation would: the broadside force is an order of magnitude larger and thrust scales with power. What it removes is the CREW, not the need to weathervane.
  • Dogged on to a mast the same ship rides out 35 m/s. THE SHIP IS SAFE IN A GALE AND HELPLESS IN A BREEZE, and every ground operation must be designed around that inversion.
  • Steady axial drag at that wind is only 36 kN, and the mast must be designed to 120 kN in any direction, a factor of 3.3. A moored airship hunts: it yaws off the wind, sails back across it, and arrives at the end of each swing with the added mass of the displaced air behind it. Design the mast to the drag figure and it fails in the first squall.

What would change the answer

A well-supported no is worth more than an optimistic yes, and it is only worth anything at all if it says where to push.

  1. 01GROUND HANDLING IS NO LONGER A BLOCKER, and vectored thrust is why

    Two people hold it broadside in 0.90 m/s, which has not changed and will not; but the ship holds ITSELF bow-on in 17 m/s, above the 6.3 m/s the US Navy would dock a ZPG-3W in with a mobile mast, two mechanical mules and eighteen trained people. A vehicle that can point into the wind and stay there does not need any of them. It still must never be held across the wind, and the mooring must still be single-point.

  2. 02BUY THE MEMBERS, DO NOT LAY THEM UP

    Pultruded tube is $262/kg against $116/kg for the fabric alone, which is 2.3 times as much and is the wrong comparison: a kilogram of finished laminate is that fabric plus resin plus single-use consumables plus tooling plus about an hour of hands per square metre of ply, and it arrives at 103 GPa rather than the 64 GPa a hand wet layup reaches. It removes 16 thousand hours of laminating and tooling, which is 34 percent of the build, plus the whole bagging consumables line and the oven. It is the only change in this module that improves cost, schedule and structure at once.

  3. 03MOOR IT ON WATER

    A ship that never comes ashore never needs the mooring circle, the mules or the ground crew, and it weathervanes off a bow drogue by itself. The water landing requirement is not a feature bolted on to this design, it is what replaces $7.8M of ground equipment.

  4. 04ASSEMBLE OUTDOORS AND INFLATE ONCE

    The shed exists because a bare frame will not survive weather and a finished hull cannot be held in a breeze. A build sequenced so that the hull is never both complete and uninflated, in a place with a reliable calm season, is the only route that does not need the building. It is a serious plan with a serious weather risk, and it is the one worth studying.

  5. 05BUILD IT SMALLER

    Everything here scales: the shed with the cube, the labour with the surface, the handling crew with the side area. The endurance figure is what a smaller ship gives up, and the trade is in the design explorer rather than asserted here.

None of those four makes this a two-person build. The first one is still worth doing on its own terms: buying the frame members instead of laminating them is cheaper per kilogram than the fabric once the resin, the consumables, the oven and the hours are counted, and the bought tube arrives at nearly double the modulus because it was cured in a heated die at a fibre volume fraction no hand layup reaches.

The second is the reason the water landing requirement is in the brief at all. It is not a feature bolted on to a flying machine. It is what replaces $7.84M of ground equipment and the crew to work it.

The third is the only route that removes the building, and it is a real proposal with a real weather risk rather than a way of avoiding the conclusion. The fourth is the honest one: everything on this page scales, and the endurance figure is what a smaller ship gives up.