Military Add-ons for P3D

Building a UH-1Y for Prepar3D: Geometry Comparison

Published:

Updated:

UH-1Y for P3D

Development pass 01 · October 31, 2026

The project begins with a simple low-poly UH-1Y base, something that was initially going to be used as static model to sit on the deck of an amphibious ship, as part of our original US NAVY fleet traffic project. This is how I went about the CH-53E AI aircraft, which I now have flying in the sim.

While it looked the part resting on a ramp, it was essentially a statue—featuring no turning rotors, non-functional landing gear, and a static cargo ramp. Transforming that basic placeholder into a fully animated, operationally viable asset required a complete overhaul from the ground up.

The goal is to develop it into a more detailed, flyable aircraft for Prepar3D v4 and later, with a proper UH-1Y cockpit, updated textures, working displays, opening doors and articulated gun assemblies, as well as interactive gauges that track the location of ships and make use of some other new functionality that are still in development.

So obviously we want to do our own work here. Starting from scratch is one way, and this is very close to doing that, but here we are using AI to plot the dimensions or the general shape and geometry, using existing models as a guide. I chose two older payware models that I already had in the sim hangar, and asked AI to get started plotting data. The objective is create something that is original, and not just copy parts.

This first geometry pass establishes a repeatable comparison process, which I will touch on later. It is a development checkpoint: the aircraft is still visibly primitive, and the new door and gun work has not yet been exported to the simulator.

Choosing the references

Area 51’s UH-1Y is the main guide for the airframe’s shape and variant-specific layout. CeraSim’s unmodified Bell 412 EP provides a second reference for applicable construction and mechanical details; its fuselage shape is not the target for the UH-1Y. Maryadi’s MV-22B remains the benchmark for the eventual cockpit textures, display clarity and functional depth.

For this pass, the geometry references retain their original scale and orientation. A provisional translation aligns the rotor mast landmarks. Local alignment against the nose, cabin sill, skids and tail is still required before final shape decisions.

Comparing the shapes

The comparison uses matching orthographic views, plane intersections at 14 longitudinal stations, and one-way nearest-surface measurements from the base to the Area 51 reference. The measurements help identify regions that need inspection. They are not measurements of real-aircraft accuracy, and the section envelopes can include fittings, skids and stabilizers.

The tail and skid regions show larger upper-end deviations in this initial scan. Those differences must be inspected in context: provisional registration, missing details and differences between the reference models can all contribute.

Side views

Our low-poly base — side view.

Area 51 UH-1Y reference — side view.

CeraSim Bell 412 EP reference — side view.

Top views

Our low-poly base — top view.
Area 51 UH-1Y reference — top view.
CeraSim Bell 412 EP reference — top view.

Front views

Our low-poly base — front view.
Area 51 UH-1Y reference — front view.
CeraSim Bell 412 EP reference — front view.

Three-quarter views

Our low-poly base — three-quarter view.
Area 51 UH-1Y reference — three-quarter view.
CeraSim Bell 412 EP reference — three-quarter view.

Solid-shaded geometry views do not represent final textures, transparency or simulator appearance.

Initial scan measurements

RegionMedian distance95th percentile
Nose and windshield7.0 cm16.1 cm
Cabin and doors10.0 cm37.1 cm
Engines and transmission10.0 cm25.2 cm
Tail boom12.6 cm39.5 cm
Tail fin region13.4 cm54.2 cm
Skids and steps10.0 cm56.0 cm

Distances are diagnostic comparisons to A51 after provisional mast alignment, not real-aircraft tolerances.

Preparing the moving parts

Four panels have been separated from the base: two sliding cabin doors and two hinged cockpit doors. Their provisional outlines follow projected Area 51 door shapes. Each has a separate controller, initial thickness and opening reveals. Closed-state surface checks passed after correcting a cutting issue.

Two original six-barrel minigun blockouts establish the assembly structure: receivers, cradles, mounts, ammunition boxes and independently controlled swivel, elevation and barrel rotation. They are layout and articulation studies, not finished minigun replicas.

Door frames, tracks, hinges, seals, separate glazing, interior surfaces, gun feeds and collision clearance all remain unfinished. Opening the doors now exposes that unfinished interior work.

Four separated door panels in the closed position.
Provisional sliding and hinged door movement test.
Six-barrel gun blockouts with independent articulation controls.

What comes next

We continue to iterate.

The initial task I gave AI was to come up with a repeatable compare → rebuild → verify process for each section, with a saved checkpoint after each completed pass.

For every section, we will:

  1. Isolate the corresponding geometry from our base model and our two guide models.
  2. Align the references using local landmarks and measured dimensions.
  3. Produce matching views, wireframes, cross-sections, and overlays to identify differences in shape and detail.
  4. Rebuild our geometry, then check its silhouette, clearances, shading, UVs, and polygon cost.
  5. Export and verify the compiled model before moving on.

The next pass will refine the nose, windshield and cabin shell against the UH-1Y reference. Door apertures will then be refitted to the improved contours before detailing their frames and mechanisms. The simulator load-test prototype remains separate from these unfinished geometry checkpoints.

Stay Tuned for Updates

Who knows how effective this approach might be. Based on recent experience with my other CH-53E model, I believe it will certainly result in a decent, low-poly AI, if nothing else. But I would be disappointed in that. I’d like to think we can exceed what we already have, which is just a 20-year old FSX model – I really want to create something on the level of Miryadi’s MV-22 in terms of quality freeware.

Quick note: Already got this ugly block flying in the sim. More to come.

Latest Posts

  • Building a UH-1Y for Prepar3D: Geometry Comparison

    Building a UH-1Y for Prepar3D: Geometry Comparison

    Development pass 01 · October 31, 2026 The project begins with a simple low-poly UH-1Y base, something that was initially going to be used as static model to sit on the deck of an amphibious ship, as part of our original US NAVY fleet traffic project. This is how I went about the CH-53E AI…

    Read more →

  • Bringing a CH-53E model to Life in Blender

    Bringing a CH-53E model to Life in Blender

    Could AI tools generate a working flying AI in P3D? A static CH-53E Super Stallion model was sitting in my files from an earlier project. It looked the part on a ramp, but it was a statue: no turning rotors, no gear, no cargo ramp. Meanwhile, the AI CH-53Es flying around my Okinawa were the…

    Read more →

  • Building Okinawa for P3D:

    Building Okinawa for P3D:

    Helicopters, Mesh, Forests and Boots on the Ground Why Okinawa, and why this way Over the past few weeks I have turned Okinawa in Prepar3D v4 into something I always wanted: a living Marine Corps island. Helicopters launch from Futenma and fly real sorties. Landing zones sit in proper jungle clearings, dozens of them. The…

    Read more →