Power levels

Power levels from acquisition parameters are represented using the Power type, which handles both Watts (W) and dB attenuation units with automatic conversion.

Creating Power objects

Power values must be created with an explicit unit specification:

# From dB attenuation
p1 = Power(30.0, :dB)

# From Watts
p2 = Power(0.001, :W)

# Works with any numeric type
p3 = Power(20, :dB)

Accessing from acqus data

Power levels are automatically parsed when loading data:

spec = loadnmr("experiment")

# Get power level (returns Power object)
p = acqus(spec, :pl, 1)  # First power level

Accessing values

Use db() and watts() to retrieve values in either unit:

p = Power(20.0, :dB)

db(p)     # 20.0 (dB)
watts(p)  # 0.01 (W)

Conversion formulas:

\[\mathrm{dB} = -10 \log_{10}(\mathrm{watts})\]

\[\mathrm{watts} = 10^{-\mathrm{dB}/10}\]

Zero power handling

Zero watts is handled specially to avoid logarithm issues:

p_zero = Power(0.0, :W)
db(p_zero)     # 120.0 (very high attenuation)
watts(p_zero)  # ≈ 1e-12

This is relevant when parsing acqus files where unused power levels may be zero.

Converting to RF field strength

The hz() function converts power to radiofrequency field strength in Hz using calibration data.

Using a reference RF strength

# Known calibration: 25 kHz at 30 dB
ref_power = Power(30.0, :dB)
ref_hz = 25000.0

# Calculate RF at different power
test_power = Power(36.0, :dB)
rf = hz(test_power, ref_power, ref_hz)

Conversion formula:

\[\mathrm{Hz} = \mathrm{Hz_{ref}} \times 10^{-\Delta\mathrm{dB}/20}\]

where $\Delta\mathrm{dB} = \mathrm{dB} - \mathrm{dB_{ref}}$.

Using pulse calibration

# Known: 10 μs pulse gives 90° at 30 dB
ref_power = Power(30.0, :dB)
pulse_length = 10.0    # μs
flip_angle = 90.0      # degrees

# Calculate RF at different power
test_power = Power(33.0, :dB)
rf = hz(test_power, ref_power, pulse_length, flip_angle)

This first calculates the reference RF strength:

\[\mathrm{Hz_{ref}} = \frac{\theta}{360 \times t}\]

where $\theta$ is the flip angle (degrees) and $t$ is the pulse length (seconds), then applies the power scaling formula above.