Community Browser
Showing results for 
Search instead for 
Did you mean: 

Re: Struggling With Measurement System Setup?

‎01-28-2016 11:30 AM
‎01-28-2016 11:30 AM

Today’s post is the first in a new series exploring areas of focus and innovation for NI data acquisition software.

Shelley headshot.jpg

Today’s Featured Author

Jordan Griffith believes that setting up and taking measurements should be as easy as ABC. As a product manager on the software marketing team at NI, Jordan spends most of her time making sure our data acquisition software meets the evolving needs of users.

It’s the afternoon of your big design review deadline. You are responsible for providing the data that validates your latest decisions. You believe your design is solid – but now you need the data to prove it, and you’re feeling the time crunch. You’ve meticulously planned out your data acquisition system to get the data that you need. All the components have arrived, your sensors are connected, and you’re ready to go. As the data starts to roll in, your heart sinks. It’s noisy, it’s unreliable, and it doesn’t look anything like you anticipated.

If you’ve built a measurement system before, you know the first time you set up your system you likely won’t get the results you need. Setting up a data acquisition system is complicated. Depending on your measurement, you need to be intimately familiar with hardware specifications, sensor calibration, and wiring schemes, not to mention the time it takes to get the software set up correctly to measure your signal. A perfect DAQ setup on the first try is like getting that perfect first compile on that few hundred lines of code you’ve written – we all dream about it, but rarely does that dream prove to be a reality. Debugging your data acquisition system is often the most time consuming part of getting your data. When the data doesn’t look right, the uncertainties start creeping in. Was that measurement supposed to be reference single ended? Did I provide the right excitation voltage to my sensor? What was that scaling co-efficient again?


Figure 1: Systems are complex, and data acquisition software isn’t making things any easier.

Engineers have long accepted challenges in system setup, and the justifications are abundant. “Debugging is part of the process.” “Yeah, but does anyone get their measurement right the first time?” “It’s not the system’s fault, I should have known better.” It’s time for those excuses to come to an end. Data acquisition systems are complicated, but the software that comes with DAQ systems doesn’t make the engineer’s job any easier. Today, the software packaged with data acquisition devices does little to help users understand and document system connections such as wires between sensors, calibration options, and grounding schemes.

It’s time for the software provided with data acquisition devices to help users overcome these complexities. Better data acquisition software could reduce system setup complexity through improved system visualization, recommendations for correct wiring, and better checks for channel configuration. To keep pace with the growing demand for data, it’s time to demand more from the software we use to acquire it.

Let us know what your biggest system set up struggles are in the comments!


This may be what we call "to come crashing through an open door" in Norway:

When applying scaling and offset (applying calibration constants) to raw data we developed a simple subVI that utilizes the suitable native LV function. But because we use the waveform data type, we're able to add "documentation" to the data at an early stage. Waveform attributes containing scale, offset, unit, (and you could add more here) are kept along with the data through the code, and are written by default to the TDMS files used for storage. This allows the data user to easily check what has been done to the data, and it's also easy to do corrections because you're able to back-calculate to raw values.