Firefighter robot

Pump inside the torso

Design review for the hardware build. Written on 13 October 2026.

The team wants the pump inside the torso, for a lower centre of mass and better balance. The reservoir is ours to design, and we can 3D print it. This article proposes two printed tanks, R1 and R2, gives their CAD files, and compares them in MuJoCo with the earlier concepts A_real (water gun on the pan-tilt) and B_real (backpack with a 12 V diaphragm pump).

Note: Status: planned for the parts and tested in simulation for the results. Nobody has printed a tank or measured anything on the robot. All masses of added parts are estimates.
Warning: The free space comes from the Berkeley Humanoid Lite CAD mesh. Our torso can differ. Check the measurements in section 6.1 on the real robot before you print.

Summary

OptionWaterAdd-on mass, fullRobot CoM heightOne-leg pitch torqueWalk speed for 0.30 m/sBackward push, walking
A_real: gun on the pan-tilt≈ 0.45 L1.54 kg0.425 m5.2 Nm0.29 m/s7.9 N s
B_real: backpack0.95 L2.76 kg0.429 m3.9 Nm0.20 m/s4.5 N s
R1: tank inside the torso0.63 L1.65 kg0.413 m5.7 Nm0.30 m/s7.9 N s
R2: side tanks0.73 L1.92 kg0.410 m5.8 Nm0.29 m/s9.8 N s
The add-on mass includes the pan-tilt, the nozzle or gun, and the splash guard. The bare robot has its centre of mass at 0.362 m. The push value is the worst of three gait phases at 6 Nm. All values are tested in simulation.

Free space in the torso tested in simulation

We measured the torso mesh of the Berkeley robot. Positions are in mm in the robot frame: x forward, y left, z up from the floor. The top face of the torso frame (the 2020 top rails) is at z = 791. The right column gives the depth under that top face, which is easier to measure on the robot.

RegionPosition (mm)Depth under the frame top (mm)Content
Battery (6S LiPo)x −30 to 33, y −66 to 89, z 583 to 633158 to 208Low in the torso. Its end sticks out of the left side to y = 89.
Mini PCx −52 to 51, y ±58, z 639 to 679112 to 152On the front and rear shelves.
USB boardx 14 to 40, y ±67, z 685 to 71180 to 106Front, above the PC.
Shoulder brackets|y| ≥ 66.7, z 681 to 7910 to 110Both sides of the upper torso.
Pan base of the ST3215 pan-tiltx −78 to 2, y ±40, z 762 to 7910 to 29Hangs under the deck.
Free: upper torsox ±48 (±62 between the posts), y ±63, z 712 to 7865 to 79R1 goes here.
Free: torso sides|y| 77 to 124, x ±66, z 590 to 678113 to 201R2 goes here. Above the hip motors (top at z 581), under the shoulder brackets.

The gaps beside the battery are 17 mm wide, and the pump is 25 mm. So no tank and no pump fits low inside the torso.

Pump

Note: The request named “Centrifugal Brushless Motor Micropumps for Liquids – for 3.2mm ID tubing”. We did not find that shop page. This article assumes the TCS Micropumps M400K-V. Check the part number before you print.
ItemValue
ModelTCS Micropumps M400K-V, brushless centrifugal, parallel ports for 3.2 mm bore tubing
Size, mass40 × 26 × 25 mm, 32 to 34 g
At 12 V5.8 W (about 0.5 A), 380 mbar maximum pressure, 1150 mL/min free flow
At 6 V1.2 W, 86 mbar, 500 mL/min
ProtectionIP45 standard, IP67 waterproof version
PrimingNot self-priming. The “-SUB” version is the self-priming, submersible one.
Source: TCS M400 range page, read on 13 October 2026.

A centrifugal pump cannot pull water up into itself. It must sit below the water, so that water flows into it by gravity. This rule decides where the pump can go.

Tank options

R1: tank inside the torso planned

Warning: R1 puts water above the mini PC. The drip tray catches pump drips only. A leak from the tank floor drips onto the PC. Leak-test the tank and add a splash cover over the PC.

R2: side tanks planned

Warning: Over the full joint range, the folded foot touches the rear end of each R2 tank. Set a software knee limit of 1.4 rad, as for the backpack. With this limit the clearance is 47 mm. The tube across the front is not checked against the legs.

Simulation results tested in simulation

Balance and leg torque

We used the same slow poses as in the earlier payload rounds: standing, crouching and standing on one leg. One-leg stance happens in every step. The walking policy was trained with a 6 Nm torque limit.

Walking and pushes

We ran the shipped Berkeley walking policy in our MuJoCo loop, with the training sensor noise. For the push test, a force acts on the torso for 0.1 s while the robot stands or walks at 0.3 m/s. A search finds the largest push with no fall in the next 3 s. The push comes at three gait phases, and the table keeps the worst.

ModelStanding: forwardbackleftrightWalking: forwardbackleftright
bare robot7.54.53.83.86.07.91.92.6
A_real9.44.54.14.18.67.92.63.8
B_real, full5.63.44.54.58.64.52.62.6
R1, full8.64.94.14.17.57.92.63.4
R1, half9.05.24.54.17.97.12.63.4
R2, full9.05.24.14.17.19.82.63.4
R2, half8.65.24.14.17.58.22.63.0
Largest push survived, in N s (0.1 s on the torso, 6 Nm). The search resolution is 0.375 N s, so differences under 0.4 N s mean nothing.
Bar charts of the largest push survived in four directions, standing and walking, for each model.
Largest push the policy survives, standing (left) and walking at 0.3 m/s (right).

In the MuJoCo videos, the red and blue boxes stand for the masses of the added parts.

Water slosh

Water in a part-full tank moves. We modelled the first sloshing mode of each tank as a mass on springs, with the standard model for a rectangular tank (NASA SP-106). The worst case is a half-full tank.

Half-full tankSlosh share of the waterFrequency fore-aft / sidewaysPeak travel while walking and turning
R1, no baffles71 %2.0 / 1.9 Hz31 / 33 mm
R1, CAD baffles39 %5.1 / 4.3 Hz8 / 10 mm
R2, no baffles60 %2.2 / 4.4 Hz24 / 7 mm
R2, CAD baffles24 %5.0 / 4.4 Hz3 / 3 mm
The damping ratios (0.01 without baffles, 0.05 with baffles) are guesses. The model has one sloshing mode.

Jet

Range, exit speed and flow against nozzle diameter. Range and speed fall fast as the nozzle gets wider.
Range, exit speed and flow of the M400K-V at 12 V.

We assumed a straight-line pump curve between the two data points, then took off the lift to the nozzle (0.2 m) and the friction in 0.6 m of 3.2 mm tube.

NozzleFlowExit speedRange to the floor, no dragSpray time, R1 full
1.0 mm0.31 L/min6.6 m/s5.3 m2.0 min
1.5 mm0.53 L/min5.0 m/s3.3 m1.2 min
2.0 mm0.67 L/min3.5 m/s2.0 m0.9 min
3.0 mm0.77 L/min1.8 m/s0.8 m0.8 min

Hardware review

What to check on the robot

Measure from the top face of the torso frame (the 2020 top rails) downward. Compare with the CAD values.

  1. For R1, the space above the mini PC must be free down to 110 mm under the frame top. The tank floor is 77 mm under the top, and the pump tray reaches 109.5 mm.
  2. For R1, the free width between the side parts of the upper torso must be at least 126 mm (y ±63 mm), and the free length between the top rails at least 96 mm.
  3. For R1, check where the USB board and the cables sit. The CAD has the board at the front, 80 to 106 mm under the top, and a cable channel at the front right.
  4. For R1, check that the pan base and pan servo hang no more than 29 mm under the deck.
  5. For R2, the side faces of the posts must be free from 113 to 199 mm under the frame top, with free T-nut slots at 124 and 188 mm.
  6. For R2, check how far the battery end sticks out of the left side. The CAD pocket clears 89 mm from the centre line.
  7. Confirm the pump part number and its port layout. The CAD assumes the M400K-V with both ports on one end.

Printed parts

PETG at 1.27 g/cm³, printed solid (100 % infill or 6 perimeters), so that the walls hold water. Wall 2.5 mm, baffles 2.0 mm.

PartPrint piecesMass (g)Size (mm)
R1 tanktwo halves, cut at y = 0, 139 g and 132 g271140 × 126 × 110
R1 deck (replaces the deck of the pan-tilt)one, flat≈ 91 at the deck settings158 × 130 × 6
R2 tank, eachtub 133 g and inner plate 33 g167141 × 49 × 100
R2 pump holderone2028 × 45 × 32
All print pieces fit a 220 × 220 × 250 mm bed. Each STL is one closed body.
OptionWaterEmpty systemFull system
R10.632 L0.349 kg0.981 kg
R20.725 L0.526 kg1.251 kg
Tank print, pump, screws and tubing. The screws and tubing are guesses (45 g for R1, 140 g for R2).

Bought parts

ItemR1R2Note
TCS M400K-V pump11Or the part the team ordered
Brass hose barb, M5 thread, for 3.2 mm bore tubing12Tank outlets. Use thread sealant.
Silicone tubing, 3.2 mm bore≈ 0.6 m≈ 1 mPump feed and jet hose
Tee for 3.2 mm tubing–1Joins the two tanks
Tapered silicone stopper1 for a 19 mm bore2 for a 16 mm boreDrill a small vent hole
M4 heat-set insert (M4 × 8.1, hole 5.6 mm) and M4 screw through the deck4–Screw length: deck 6 mm plus the insert depth
M5 × 55 screw, washer, drop-in T-nut for 2020–8Into the side slots of the posts
Two-part epoxyyesyesJoins the print pieces
Drain tube, 6 mm≈ 0.3 m–From the drip tray out of the torso

Print and assembly

  1. Print the R1 halves on their cut face. The notch for the pan base is a 27 mm bridge in this orientation. Use supports there if your printer sags.
  2. Glue the R1 halves with epoxy along the cut face. The baffles are cut in half too, so they glue at the same time.
  3. Print the R2 tubs on their outer face, and the inner plates flat. Glue each plate on its tub with epoxy.
  4. Fill each tank with water and let it stand for one hour on paper. If it leaks, coat the inside with thin epoxy and test again.
  5. R1: press the four M4 inserts into the pillars. Tap the outlet M5 and screw in the barb with thread sealant.
  6. R1: put the pump in the tray and connect the outlet barb to the pump inlet with a short tube. Lower the tank into the torso from the top. Bolt the deck to the rails and the tank under the deck.
  7. R2: drop eight M5 T-nuts into the side slots of the posts. Bolt each tank with four M5 × 55 screws.
  8. Fill the tank and wait until water reaches the pump before you switch it on. Check the TCS manual for dry running.

Downloads

torso_tank_cad.zip holds the STEP and STL files of every printed part, the assembled tanks, simple blocks for the pump, the check log and the CAD notes. All parts are drawn in the robot frame in mm, so the STL files stack on the robot without a transform.

FileWhat
r1_half_left, r1_half_rightR1 print pieces
r1_tankR1 assembled, for fit checks
r1_deckDeck of the pan-tilt with the four tank holes
r2_tub, r2_inner_plateR2 print pieces (left tank; mirror them for the right tank)
r2_tank_left, r2_tank_rightR2 assembled, for fit checks
r2_pump_holderPump holder for R2
ref_pump_r1, ref_pump_r240 × 26 × 25 mm blocks where the pump sits

Recommendation

  1. Choose R1 if the pump must be inside the torso. It feeds the pump by gravity, needs no knee limit, adds no width, and keeps the tank inside the frame.
  2. Before R1 goes in the robot, leak-test it, add a splash cover over the mini PC and route the drip tray drain out of the torso.
  3. Choose R2 if water above the computer is not acceptable. It needs the 1.4 rad knee limit, a pump outside the torso and a tube across the front.
  4. Order a 1.0 mm and a 1.5 mm nozzle, and measure the jet speed before the first fire test.
Tip: When the course model of the robot (“Robot XML v1” on the course site) is out, we will check the free space again and rerun the simulations with it.