Extend "Make a measurement" duration?

[workingstorage]

I'd like to let the UMK sit for a couple of hours outside to capture the exterior noise for a period of time, and the "Make a measurement" option only goes to around 70 seconds.

Is there a way to extend this time? And write the output file to a larger drive?

The ultimate goal would be to review the measurement for peaks (when especially loud vehicles are passing by my house) and extract this selected data out of the file.

I know it's not the primary use case for this software, but it's a growing one, given how loud the world is getting these days. Would appreciate any thoughts.

Thanks!

 

[John Mulcahy]

You can use the Logger on the SPL meter for that.

 

[JHBrandt]

Rob, we need the SPL and Frequency at full bandwidth. As far as I know, the only way to do that is with an extended sweep without using a sound source(speakers)

 

[John Mulcahy]

Rob, we need the SPL and Frequency at full bandwidth. As far as I know, the only way to do that is with an extended sweep without using a sound source(speakers)

That's not at all reliable as the lack of anything derived from the stimulus in the response can result in misalignments and silent parts of capture being part of the result. The RTA is the tool for capturing spectral content, either live or by analysing sections of a recording. Easiest for longer periods would probably be to capture a recording with Audacity and use Audacity's spectrogram to look at the spectral content over time.

 

[JHBrandt]

John, on my resources page I have a document detailing how to do this. It has been very reliable as long as the UMIK-1 is used with cal file. Please check it out and let me know your thoughts. THANKS!
Meh... I can't put a link here, your forum doesn't allow links. Go to my website then the resources page. Look at the third document down, "Assess the Environment".
Cheers,
John Brandt

 

[John Mulcahy]

Unfortunately the process can't always give reliable results. The sweep result is generated in essence by dividing the FFT of the captured data by the FFT of the stimulus and doing an inverse FFT to get the impulse response. Where the capture has some correlation with the stimulus the impulse response will be meaningful and have a peak corresponding to the direct sound. A region around the peak is windowed and the FFT of that windowed region produces the plot. When the capture is unrelated to the stimulus the peak could be anywhere, including in parts of the impulse response that are affected by zero padding of FFTs during the calculation and so include chunks of digital silence and processing artefacts. Windowing those regions would not produce a response that reflects the environmental noise.

The RTA is a better way of capturing background noise levels. If 'forever' averaging is used the peak trace will capture the maxima from the time it is reset until the capture is stopped, that would probably give a much more accurate picture.

 

[JHBrandt]

Thank you for this.

How can we 'save' the RTA in a format that can be exported? I need SPL per 1/3 octave band.

 

[John Mulcahy]

The peak or current RTA trace can be saved as a measurement by using the save buttons above the graph or the Alt+P or Alt+S shortcuts respectively, those saved measurements can be exported as normal. The RTA could be set in 1/3 octave mode before making the capture, with the 'Appearance' option to "Use bars on RTA" to simplify reading the levels.

 

[JHBrandt]

Thank you, John.

I'll test that and get back to you.
Note: we've been making these site measurements this way (sweep with no speakers) for a while and they work wonderfully. Perhaps we can talk about this function over zoom?
Cheers,