mobius-units
Fundamental constants and unit conversions from the eigenvalue tower — one measurement derives them all.
from mobius_units import alpha_inv, m_e, G, l_P
alpha_inv # 137.03599917656376 (tower-derived, not CODATA lookup)
m_e # 9.109383714e-31 kg (from tower alpha + R_inf)
G # 6.67436e-11 m^3 kg^-1 s^-2 (tower-derived, 0.37 sigma)
l_P # 1.61626e-35 m (tower-derived)
The Problem
Existing unit libraries (pint, astropy.units, scipy.constants) carry 20+ empirical conversion factors from CODATA lookup tables. When CODATA updates, the tables must be refreshed. No algebraic relationship connects the constants — they are independent numbers from independent measurements.
The Fix
Three algebraic equations produce three dimensionless quantities with zero measured inputs:
| Equation | Produces | Agreement with experiment |
|---|---|---|
a^-1 + S*a = 4*pi^3 + pi^2 + pi |
alpha (fine structure constant) | 0.3 sigma |
6*pi^5 * (1 + a^2/(2*sqrt(2)) - (22/27)*a^4) |
mu (proton-electron mass ratio) | 0.03 sigma |
ln(mP/mp) = 14*pi + pi^2*sqrt(a)/28 |
eta (Planck hierarchy) | 4.2 ppm |
Combined with five SI-exact constants (c, h, e, k_B, N_A) and one measurement (the Rydberg constant R_inf), these produce the entire CODATA table.
Install
pip install mobius-units
Requires Python >= 3.9. One dependency: mobius-constant.
Constants
Every alpha-dependent constant uses the tower value, not a CODATA lookup.
Dimensionless (tower-derived)
| Name | Symbol | Value |
|---|---|---|
alpha_inv |
1/alpha | 137.035999177 |
alpha |
alpha | 7.2973526e-3 |
mu |
mp/me | 1836.15267343 |
eta |
ln(mP/mp) | 44.0124 |
Atomic scale (tower alpha + R_inf)
| Name | Symbol | Value | alpha-dep |
|---|---|---|---|
m_e |
electron mass | 9.109e-31 kg | alpha^-2 |
m_p |
proton mass | 1.673e-27 kg | alpha^-2 |
a_0 |
Bohr radius | 5.292e-11 m | alpha^1 |
r_e |
classical e- radius | 2.818e-15 m | alpha^3 |
lambda_C |
Compton wavelength | 3.862e-13 m | alpha^2 |
mu_B |
Bohr magneton | 9.274e-24 J/T | alpha^2 |
E_h |
Hartree energy | 4.360e-18 J | alpha^0 (exact) |
sigma_T |
Thomson cross-section | 6.652e-29 m^2 | alpha^6 |
Planck scale (tower hierarchy)
| Name | Symbol | Value | Agreement |
|---|---|---|---|
G |
gravitational constant | 6.674e-11 | 0.37 sigma |
m_P |
Planck mass | 2.176e-8 kg | 0.36 sigma |
l_P |
Planck length | 1.616e-35 m | 0.38 sigma |
t_P |
Planck time | 5.391e-44 s | 0.37 sigma |
T_P |
Planck temperature | 1.417e32 K | 0.36 sigma |
Unit Systems
Three natural unit systems, all algebraically connected:
from mobius_units import AtomicUnits, PlanckUnits, ParticleUnits
# Atomic -> SI
AtomicUnits.to_si(1.0, 'energy') # 1 Hartree in joules
AtomicUnits.to_si(1.0, 'length') # 1 Bohr in metres
AtomicUnits.to_si(1.0, 'time') # 1 atomic time in seconds
# SI -> Atomic
AtomicUnits.from_si(9.11e-31, 'mass') # electron mass in atomic units (= 1)
# Planck -> SI
PlanckUnits.to_si(1.0, 'length') # 1 Planck length in metres
PlanckUnits.to_si(1.0, 'temperature') # 1 Planck temperature in kelvin
# Particle physics (SI-exact, no tower needed)
ParticleUnits.to_si(125.0, 'energy') # 125 GeV in joules
# Cross-system
from mobius_units import atomic_to_planck
atomic_to_planck(1.0, 'energy') # 1 Hartree in Planck energy units
The Alpha Source
All alpha-dependent constants trace to mobius-constant:
from mobius_constant import Alpha_inv
float(Alpha_inv) # 137.03599917656376
This is the tower-derived value, not the CODATA adjusted average. It sits 0.3 sigma from the best measurement in physics (Fan et al., 2023). The ~4.5 sigma offset from CODATA 2018 adjusted constants traces entirely to the documented Cs-133 recoil anomaly.
See Also
- mobius-constant — Exact irrational constants
- One Measurement Derives Them All — companion paper (Carpenter, 2026)
License
MIT — Jay Carpenter, 2026
Metadata
Release files for mobius-units 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| mobius_units-0.1.0.tar.gz | 8.6 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| mobius_units-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 16.7 kB
Release files / mobius_units-0.1.0.tar.gz
| Download URL | mobius_units-0.1.0.tar.gz |
|---|---|
| Size | 8.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e82d7619baf026e5d98d101922c4ba96935bb7cf2595a77b26a87b8aa4fd210a
|
|
BLAKE2b-256 checksum How to use checksums |
538dee96e682d1f0db75ea6460797693f347395888d109937230b77bf99aad7e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Release files / mobius_units-0.1.0-py3-none-any.whl
| Download URL | mobius_units-0.1.0-py3-none-any.whl |
|---|---|
| Size | 8.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ab3f882f3e2620c7eba79a5a116d15000f40d404707826d9bd0bd7ee04319496
|
|
BLAKE2b-256 checksum How to use checksums |
caa49943765b39e42d9fc2fec890e10124b946b3fcec22a34f2bb2201d5094c4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|