Sound waveform to generate square waveform

[RishiDivakar]

Hi All,

I have a requirement where I need to convert an automotive chime tone of 1.2 seconds to a pulse waveform. ie, from the starting of the chime to the ending of the chime (1.2 sec) the waveform should be high. However, since the waveform is having different sine components, I'm not able to get the exact waveform 

I'm not much familiar with signal processing, only aware of the basics. Can anyone please help me how to resolve this?

[GerdW]

Hi Rishi,

convert an automotive chime tone of 1.2 seconds to a pulse waveform. ie, from the starting of the chime to the ending of the chime (1.2 sec) the waveform should be high.

Simple solution, based on this desription: create a waveform signal, containing "high" values for 1.2s duration!

Multiply the chime waveform with zero, then add 1:

[RishiDivakar]

Hi Gerdw,

I forgot to mention one thing. I will be continuously acquiring the data and the duration of the tone is 1.2 sec not of the acquisition. Means, my total acquisition time might be more than 20 seconds but, the chime duration where chime is present it will be 1.2 seconds. Since it's a sine burst, the signal is going to zero multiple times, and because of that I'm not able to get a perfect pulse of duration 1.2 seconds.

[GerdW]

Hi Rishi,

I forgot to mention one thing.

It always helps to provide as much information as possible.

(It helps a lot, when you want to develop an algorithm!)

You only need to detect the start of your chime signal. Once you know the start you can simply generate your 1.2s HIGH signal…

[RishiDivakar]

Hi GerdW,

My chime is of 1.2 seconds. However, My test system doesn't know if it is of 1.2 seconds duration. It needs to identify whether the duration of the chime is 1.2 seconds. I have used the peak detector to identify the peaks and convert to a pulse waveform. But I don't think this is a conventional method. Attached is the measurement file and vi.

[RishiDivakar]

Plus, if the chime repeats multiple times, the vi will not generate pulses properly, it will generate a pulse of long duration.

[GerdW]

Hi Rishi,

looking at this signal I would detect if the input value is > 1 (as your noise seems to stay below 0.1).

Whenever an input pulse is detected then the next 1.4ms of the output signal is set to HIGH as your bursts seem to last for ~0.6ms followed by a gap of ~0.7ms… (You may adapt those numbers to your exact requirements.)

if the chime repeats multiple times, the vi will not generate pulses properly, it will generate a pulse of long duration.

When the chimes repeats very fast, then I would also expect a longer HIGH pulse output…

With every message you provide some new information/requirement specs: why don't you provide them right from the start?

Btw. when attaching such big files you shuld think about using a ZIP tool. Your chime file can be compressed to less then 500kB!

[RishiDivakar]

Hi GerdW,

Sorry for the trouble. I'm using my PC from my organization network which will not allow me to transfer any zip file. Also, since the automotive chimes may be repeated with different duty cycles, I assumed you might be aware of it.

[RishiDivakar]

Hi,

Passing the signal to a threshold detector (Comparison 1 or 0) and the result to an LPF will provide the envelope.