jondoesntgit/hardware

Python Wrappers for Hardware used in the Digonnet Lab at Stanford

hardware's Issues

Running `codecov` returns error "The filename, directory name, or volume label is incorrect"

This seems to be a problem with the vanilla install of codecov on windows. To solve this issue, find where codecov is stored on your local drive:

>>>import codecov
>>>codecov.__file__

Then, in __init__.py of codecov, find the line where it says root = quote(root) and comment it out.

Sampling too slowly introduces extra noise

Create an `instrument` base class that various instruments extend

Each instrument should have a base class that it inherits methods like identify and log_state from.

Furthermore, instruments should be divided into categories, like function generators, lock-in amplifiers

readme not accurate

The README states that one can invoke:

make help, make test, make upload, etc..

This is inaccurate, since make help isn't a legit target:

λ make help
"help" is not a valid recipe. Type "make" for help

I recommend either updating the README or changing the target in make.bat. Happy to send in a pull request, just lmk which one you'd prefer.

Allow a customizable LIA sensitivity during scale factor calibration

Currently, this sensitivity is hard-coded to .3, but it would be nice to make this customizable either

• through a parameter in hardware.gyros.tombstone
• an optional parameter in a gyro .json file.

Add unit tests for writing and reading each instrument

Test each instrument with units. Check to see whether conversions work out accordingly. For example, writing in millivolts, volts, hertz, kilohertz, and seeing if what you read back matches what you write.

Clean up __init__.py

There's a lot of code in here, that's very redundant. There is probably a way to consolidate a lot of this code.

Especially for two physical devices with slightly different IDs, that can be controlled by the same class.

Units for Power on Optical Power Meter?

When an instrument is initialized, log it

It would be nice if all of the print statements in __init__.py were replaced with logging statements. Also, if the state of each machine could be logged to a file as well.

MockDAQ should return white noise, flicker noise, random walk, etc...

Replace large dictionaries with enums

There are several places where large dictionaries are used:

• lia status bytes
• lia time constants
• lia sensitivities
• awg waveforms

These could be replaced by enum classes.

Add a `log_state` method to each instrument

Each instrument should have a log_state method that dumps its state (all of its properties) into the log.

Fix DAQmx Status Code -200557

Occasionaly, runing daq.read() spits out these errors.

Status Code: -200557
in function DAQmxStartTask
DAQmx Error: Specified property cannot be set while the task is running.

It doesn't trigger any exceptions, but terminates the read almost instantly. You end up getting a large zero vector.

make test not working

Did I miss a step?

λ make test
'py.test' is not recognized as an internal or external command,
operable program or batch file.

oh, looks like a hidden dependency?