UAV SILM logoUAV SILM
Inside the airframe

Where every part sits, and why

We placed each component against real cross-sections of the V4 fuselage, then balanced the aircraft from the 3D model. Pick a part in the model or in the list to see its position and the reason for it, and try the battery and tail-servo options to see how the balance moves.

Side view of the internal layout with every component at its fitted position
Drag to rotate, click a part
Selected part
Pick a part in the model or in the list below

PowerAvionics and payloadActuation and propulsion
ComponentMassx from the noseLevelWhy it sits there
Camera, Siyi A2 Mini85 g65 to 108 mmCentreline, nose windowAs far forward as it goes. It needs a forward view, and at 85 g it is the second-biggest weight ahead of the CG. The lens looks out through a window in the nose.
Airspeed sensor, ASPD-45258 g140 to 160 mmOn top of the batteryClose to the pitot tube, so the pressure tubes stay short. With a pusher propeller the nose is the only clean air.
Battery, 4S 7000 mAh590 g110 to 248 mmCentrelineA third of the total mass, so it is our main balance adjuster: 10 mm of movement shifts the CG by 3.5 mm. It sits on the centreline, where the round fuselage is widest, and the nose taper stops it going further forward.
Power board, FCHUB-A15 g262 to 298 mmLower deckDirectly behind the battery, so the thick battery leads are as short as possible. Everything else takes power from here.
Raspberry Pi 446 g262 to 347 mmUpper trayOn the upper tray under the wing root. Its first position, on a tray above the battery, hit the fuselage wall.
Flight controller, Matek H74328 g323 to 377 mmLower deckIts centre is 7 mm from the CG. The motion sensors work best close to the CG, where rotation adds the least false acceleration.
4G module, SIM7600E33 g350 to 416 mmUpper trayNext to the Pi for a short USB cable, and below the main spar. Its antennas go on the fuselage sides, away from the GPS.
ESC, 60 A28 g370 to 440 mmLower deckKeeps the battery-side leads short. The long motor wires run along the floor to the tail. It sits on standoffs with an air gap.
Wing servos, 2 × SG9018 g430 mm, in the wingIn the wing, 430 mm out to each sideIn the wing next to the ailerons. Only their leads enter the fuselage.
Tail servos, 2 × SG9018 g508 to 532 mm, or 728 to 752Centreline, side by sideTwo options under study: ahead of the wing trailing edge with 270 mm pushrods, or in the tail with short links. The CG differs by only 2.4 mm between them.
GPS and compass, M10Q8 g560 to 580 mmTopBehind the wing, so the carbon spar does not block the sky, and away from the ESC, motor and 4G module because of its compass.
Pushrods6 g532 to 803 mmCentreline to the ruddervator hornsCarry the tail servo movement to the ruddervators. At this length they need two guides each so they do not bend.
Motor, T-Motor AT232194 g795 to 827 mmInside the tail coneAs far aft as the tail cone allows. Its long shaft passes through the end face to the propeller. The cone needs cooling vents.
Propeller, 10 in pusher25 g860 mmBehind the tailPushes from behind, which keeps the nose and the camera view clear of the propeller.
Wiring and connectors70 gabout 335 mmSpread through the fuselageCounted as one mass at about 335 mm, close to the CG.

Positions are along the fuselage from the nose tip. The airframe itself is part of the balance too: about 630 g of printed shells, wing and tail (CAD masses calibrated on the weighed nose cone), the fuselage joints, the battery tray, the 7 mm carbon wing tube and the hatch magnets. Some small items are not placed yet.

An aircraft balances like a seesaw on its centre of gravity. The battery and camera at the front balance the motor, the tail and the rear half of the wing. The chart shows how hard each item pulls the balance point forward or aft: its mass times its distance from the CG.

For stability the CG has to sit ahead of the neutral point, where the extra lift from a gust acts. The gap between the two, as a share of the mean wing chord, is the static margin. Ours is about 20%, a comfortable value for first flights.

Why we moved the wing instead of adding weight

Our first balance check put the CG at 411 mm, behind the neutral point at 364 mm, so the aircraft would have been unstable. We moved the electronics forward, planned a lighter single-wall V-tail and moved the wing 20 mm aft. In CFD the wing move shifted the neutral point by 18.2 mm but moved the CG only about 5 mm, because the wing is only a quarter of the mass. Reaching the same margin with the old wing position would have needed about 80 g of weight in the nose.

Bar chart of each component's moment about the centre of gravity, from the camera at the nose to the propeller
Each item's pull about the CG. Left pulls forward, right pulls aft.

A side view hides the real constraint: the fuselage is round. We sliced the 3D model every 2 to 4 mm along its length and checked that each component's box stays at least 5 mm inside the outer skin, which leaves room for the 4 mm wall and 1 mm of clearance. Three clashes came up.

!

Raspberry Pi on a tray above the battery

It overlapped the wall by 0.8 mm. We moved it to the upper tray behind the battery, where it has 21 mm to spare.

!

Camera fully inside the nose at x 80

It did not fit, so the lens now looks out through a window in the nose. Enclosing it completely would push the camera and battery 35 mm aft and move the CG about 14 mm.

!

Motor at 814 to 846 mm

Only 1.1 mm of wall was left in the narrowing tail cone. We moved it forward to 795 to 827 mm (5.7 mm margin), with a long shaft out to the propeller.

  • The battery only fits on the centreline, where the round section is widest. Raised or lowered, its corners hit the wall.
  • Between 295 and 545 mm everything stays below 27 mm above the centreline, because the wing root and spar take the space above.
  • Tightest fits after the changes, apart from the camera in its window: motor 5.7 mm, battery 5.9 mm, airspeed sensor 6.6 mm, ESC 7.4 mm.

We plan to print the trays in PETG and keep LW-PLA for the skins. Foamed LW-PLA is brittle in thin plates, crushes around screws and inserts, and softens at around 55 to 60 °C.

  • Battery tray: PETG, a 2 mm floor with ribs and slots for the 11 mm slide, held with a strap and Velcro. Wide glued flanges spread a crash load of about 100 N into the fuselage.
  • Electronics: one removable two-level PETG sled, wired and tested on the bench, then slid in through the hatch.
  • Flight controller: on rubber grommets or a printed TPU plate, so motor vibration does not reach its motion sensors.
  • ESC: on standoffs with an air gap, away from the skin.
  • Fixings: heat-set inserts only in PETG parts; PETG bonded to LW-PLA with epoxy.
  • No carbon-filled filament near the electronics: it is slightly conductive and weakens GPS and 4G signals.

The trays add about 30 to 35 g. Most of it sits ahead of or at the CG, which moves the CG about 1 mm forward.

Side view drawing of the UAV SILM internal layout with labelled components, CG and neutral point
Side view with every component at its fitted position.
Top and side view drawing of the layout with the tail servos at 720 to 760 mm
Top and side views with the tail servos at x 720 to 760, the option under study.

SILM Buteo is in development and has not flown yet. Figures are design estimates from CFD and calculation, not flight-test data. About the student project