add corning fb plate compute volume

[xbtu2]

addded the appropriate compute_volume_from_height and compute_height_from_volume functions for Cor_96_wellplate_360ul_Fb in plates.py using standard cylinder volume calculations based on the well's dimensions

[rickwierenga]

thanks!

[BioCam]

Hi @xbtu2,

Could you please refactor the height/volume function to their new format:
using measured volume:height maps which are used for interpolation between known points, rather than geometric functions

See as an example: https://github.com/PyLabRobot/pylabrobot/blob/main/pylabrobot/resources/eppendorf/tubes.py

This increases accuracy, and makes the gathered functions easier to (1) correct, (2) extend.

# --------------------------------------------------------------------------- #
# 5 mL Eppendorf Tubes
# --------------------------------------------------------------------------- #

_eppendorf_tube_5mL_Vb_snapcap_height_to_volume_measurements = {
  0.0: 0.0,
  5.0: 1.342,
  10.0: 1.875,
  50.0: 4.975,
  100.0: 6.975,
  200.0: 9.909,
  400.0: 13.742,
  600.0: 16.242,
  800.0: 18.375,
  1000.0: 20.142,
  1200.0: 21.809,
  1500.0: 23.842,
  2000.0: 27.242,
  3000.0: 34.242,
  4000.0: 41.009,
  4500.0: 44.175,
  5000.0: 47.509,
  5500.0: 50.675,
}
_eppendorf_tube_5mL_Vb_snapcap_volume_to_height_measurements = {
  v: k for k, v in _eppendorf_tube_5mL_Vb_snapcap_height_to_volume_measurements.items()
}


def _compute_volume_from_height_eppendorf_tube_5mL_Vb_snapcap(h: float) -> float:
  """Estimate liquid volume (µL) from observed liquid height (mm)
  in the Eppendorf 5 mL V-bottom snap-cap tube,
  using piecewise linear interpolation.
  """
  if h < 0:
    raise ValueError("Height must be ≥ 0 mm.")
  if h > 55.4 * 1.05:  # well cavity height + 5% tolerance
    raise ValueError(f"Height {h} is too large for eppendorf_tube_5mL_Vb_snapcap.")

  vol_ul = interpolate_1d(
    h, data=_eppendorf_tube_5mL_Vb_snapcap_height_to_volume_measurements, bounds_handling="error"
  )
  return round(max(0.0, vol_ul), 3)


def _compute_height_from_volume_eppendorf_tube_5mL_Vb_snapcap(volume_ul: float) -> float:
  """Estimate liquid height (mm) from known liquid volume (µL)
  in the Eppendorf 5 mL V-bottom snap-cap tube,
  using piecewise linear interpolation.
  """
  if volume_ul < 0:
    raise ValueError(f"Volume must be ≥ 0 µL; got {volume_ul} µL")

  h_mm = interpolate_1d(
    volume_ul,
    data=_eppendorf_tube_5mL_Vb_snapcap_volume_to_height_measurements,
    bounds_handling="error",
  )
  return round(max(0.0, h_mm), 3)


def eppendorf_tube_5mL_Vb_snapcap(name: str) -> Tube:
  """Eppendorf cat. no.: 0030119401

  5 mL snap-cap Eppendorf tube with V-bottom.

  - Colour: transparent
  - Material: Polypropylene
  - Sterilized: No
  - Autoclavable: No
  - Chemical resistance:?
  - Thermal resistance: ?
  - Surface treatment: available as LoBind™ version
  - Sealing options: na
  - Centrifugation safety: Maximum safety and stability for centrifugation up to 25,000 x g
  - Total volume = 5_000 ul
  - URL: https://www.eppendorf.com/se-en/Products/Lab-Consumables/Lab-Tubes/EppendorfTubes-50mL-p-0030119401
  - technical drawing: ./engineering_diagrams/Eppendorf_5mL_snapcap_tube.pdf
  """
  diameter = 16.7
  return Tube(
    name=name,
    size_x=diameter,
    size_y=diameter,
    size_z=56.7,
    model=eppendorf_tube_5mL_Vb_snapcap.__name__,
    max_volume=5_000,  # units: ul
    material_z_thickness=1.2,
    bottom_type=TubeBottomType.V,
    compute_volume_from_height=_compute_volume_from_height_eppendorf_tube_5mL_Vb_snapcap,
    compute_height_from_volume=_compute_height_from_volume_eppendorf_tube_5mL_Vb_snapcap,
  )

[xbtu2]

@BioCam but the shape of the wells is cylindrical. would it make a difference?

[BioCam]

@BioCam but the shape of the wells is cylindrical. would it make a difference?

If it is perfectly cyclindral then no.

The issue is that almost nothing is perfectly cyclindral: most wells are injection molded and require at least some tapering, which would make a cylindrical fit incorrect.

But I think the main point is to use a model that enables increasing accuracy over time by providing simple means to change the data :)

For this one I'd just take 2 datapoints:
Use STAR to cLLD probing to measure height at dead/minimal volume, measure height at max volume, add 0 height 0 volume and that's enough