nf6x_eetools: NF6X's misc. electrical engineering related tools.

This Python package is a collection of miscellaneous tools I've created for my general electrical engineering related tasks. I use them in scripts and in interactive ipython shells.

EngFormatter

EngFormatter extends the standard string.Formatter class to support new Format Specification Mini Language types which are handy for component values, electrical measurements, etc.:

These two variants both format numbers with coefficients in the range of [1...1000), an SI prefix, and a specified total number of significant digits. Rounding is performed with the decimal.Decimal() class, using decimal.ROUND_HALF_UP. This handles significant digits in much the same way that HP scientific calculators do.

The top level package creates variable eformat as a shortcut to the format() method of an EngFormatter instance. It also provides the eng() function as a shortcut for interactive use.

For example:

>>> import nf6x_eetools as ee
>>> print(ee.eformat('R{:d} = {:3i}Ω', 123, 12345))
R123 = 12.3 kΩ
>>> print(ee.eformat('V1 = {:+2I}volt', 0.012345))
V1 = +12 millivolt
>>> ee.eng(12345)
'12.3 k'
>>> ee.eng(12345, 2, 'V')
'12 kV'

E3, E6, E12, E24, E48, E96, E192(value, unit='Ω')

These classes find the nearest IEC 60063:2015 standard component value to a specified ideal value. See E series of preferred numbers for reference.

When cast to a str, the approximated value will be formatted with EngFormatter using an appropriate number of significant digits for the series and appending SI units.

For example, to print the standard 1% tolerance resistor value closest to 50 ohms, along with its tolerance and nominal error from the ideal value:

>>> import nf6x_eetools as ee
>>> r = ee.E96(50)
>>> print(f'{str(r)} ±{r.tolerance():%} ({r.error():+.2%} error)')
49.9 Ω ±1% (-0.20% error)

ratio(v1, v2, maxsteps=1)

Find a pair of standard component values approximating a ratio.

I typically use this for tasks such as finding standard resistor values to use in a resistor divider setting the output voltage of a regulator.

Given two values v1 and v2, ratio() finds two nearby standard component values which most closely match the ratio of the ideal values. If the passed values are derived from Eseries, then the chosen value will be in the same series, within +/- maxsteps of the nearest standard value. If either passed value is not derived from Eseries, then E96 (±1%) values will be used.

maxsteps must be an integer >= 1.

Returns tuple of (w1, w2, error) where w1 and w2 are the new approximate values, and error is defined as:

                (w1.approx() / w2.approx())
error = 1.0  -  ---------------------------
                 (v1.exact() / v2.exact())

>>> import nf6x_eetools as ee
>>> r1_ideal = 1.23456E3
>>> r2_ideal = 6.54321E5
>>> (R1, R2, error) = ee.ratio(ee.E96(r1_ideal), ee.E96(r2_ideal))
>>> print(f'R1 = {str(R1)}')
R1 = 1.21 kΩ
>>> print(f'R2 = {str(R2)}')
R2 = 634 kΩ
>>> print(f'error = {error:+.2%}')
error = -1.15%

divider(v_in, v_out, r_total=E96(10000), maxsteps=1)

Design a resistor divider using standard values.

For the passed input voltage v_in and output voltage v_out, divider() designs a resistor divider using standard component values to produce v_out from v_in. The top and bottom resistors will add up to approximately r_total, and will be in the same series as r_total. If r_total is not an Eseries subclass, then E96 (±1%) values will be used. maxsteps determines how many standard value steps the top and bottom resistors may deviate from their initial approximations. Increasing maxsteps tends to reduce the v_out error and increase the deviation from r_total.

Returns a tuple of (r_top, r_bot, error), with error defined by:

error = (v_out_actual - v_out)/v_out

frac(value, denominator=64, rounding=0, correction=True)

Find nearest fractional dimension to provided value.

Returns a string representation of a fractional dimension approximating the provided value.

drill(diameter, margin=0, showdiam=True, table=drill_table)

Return string representing nearest American drill size. See Drill bit sizes for reference.

For a given diameter, return a string representing the nearest American twist drill size.

Four drill tables are currently defined:

For example:

>>> import nf6x_eetools as ee
>>> ee.drill(0.256)
'F (0.257 in)'
>>> ee.drill(0.256, 1)
'F (0.257 in)'
>>> ee.drill(0.256, -1)
'1/4in (0.25 in)'

Building and Installing from Source

After cloning the source code repository, build with the python build package, and then install the generated wheel file with pip. For example:

python3 -m build
python3 -m pip install dist/nf6x_eetools-1.0-py3-none-any.whl

(change the version number in the .whl path to point to the generated wheel file)

Running Tests on the Source Code

To run lint on the code, install pylint and run it from the top directory:

pylint ./src ./tests

To run regression tests, install pytest and pytest-cov, and run pytest from the top directory:

pytest