I would like to use the derivative operator to create a Math channel from measurement data. I've found the d/dx function calculates the derivative on a per sample basis. This makes result very "noisy". I tried the derivative of the average but could not get it to work.
Is it possible to specify the "window" for either of these two functions?
Thanks
John
Math Channel (derivative)
Re: Math Channel (derivative)
Hi John,
The Maths channels are calculated over the data in the waveform view.
Have you tried adjusting the timebase and number of samples?
It is possible to add a measurement on a maths channel between rulers.
Regards,
The Maths channels are calculated over the data in the waveform view.
Have you tried adjusting the timebase and number of samples?
It is possible to add a measurement on a maths channel between rulers.
Regards,
Hitesh
Software Dev. Engineer
Software Dev. Engineer

 Newbie
 Posts: 0
 Joined: Tue Mar 01, 2011 3:17 pm
Re: Math Channel (derivative)
+Hitesh,
Thank you for the reply, however I'm not sure I understand your response.
To calculate a derivative, I would normally calculate; delta voltage / delta time.
It appears the math channel uses the delta time for one sample to the next. This however becomes very noisy so I would normally average some readings then take the derivative. Something like:
Average 4 readings minus the average or the previous 4 readings / delta time for 4 readings.
If I create a math channel using the average(A) function for example, I don't know how many readings are averaged. Similarly I cannot control the delta time for the derivative function. Perhaps this is too much to ask for a realtime waver form calculation.
BTW my goal is to display acceleration given an analog signal of the movement. So whatever solution there is would have to also find the 2nd derivative.
Thanks
John
Thank you for the reply, however I'm not sure I understand your response.
To calculate a derivative, I would normally calculate; delta voltage / delta time.
It appears the math channel uses the delta time for one sample to the next. This however becomes very noisy so I would normally average some readings then take the derivative. Something like:
Average 4 readings minus the average or the previous 4 readings / delta time for 4 readings.
If I create a math channel using the average(A) function for example, I don't know how many readings are averaged. Similarly I cannot control the delta time for the derivative function. Perhaps this is too much to ask for a realtime waver form calculation.
BTW my goal is to display acceleration given an analog signal of the movement. So whatever solution there is would have to also find the 2nd derivative.
Thanks
John
Re: Math Channel (derivative)
Hi John,
Apologies for any confusion.
The average is actually taken across the number of waveforms rather than across time. One way to see this is if a white noise signal is fed into Channel A  it will decrease as the number of waveforms captured increases.
I will need to do some further tests in relation to the derivative calculation and will post an update when I have obtained some further information.
It is possible to calculate the derivative of a derivative using the Maths channels.
Are your calculations over a single waveform? Which PicoScope device are you using?
Thanks,
Apologies for any confusion.
The average is actually taken across the number of waveforms rather than across time. One way to see this is if a white noise signal is fed into Channel A  it will decrease as the number of waveforms captured increases.
I will need to do some further tests in relation to the derivative calculation and will post an update when I have obtained some further information.
It is possible to calculate the derivative of a derivative using the Maths channels.
Are your calculations over a single waveform? Which PicoScope device are you using?
Thanks,
Hitesh
Software Dev. Engineer
Software Dev. Engineer
Re: Math Channel (derivative)
Hi John,
With regards to the derivative, if you have PicoScope 6.6.50/57 or an early version of PicoScope 6.7 Beta, you can apply resolution enhancement on the channel and then create a maths channel to calculate the derivative.
If you reduce the number of samples on the original waveform, then you should get a less noisy derivative.
I hope this helps.
With regards to the derivative, if you have PicoScope 6.6.50/57 or an early version of PicoScope 6.7 Beta, you can apply resolution enhancement on the channel and then create a maths channel to calculate the derivative.
If you reduce the number of samples on the original waveform, then you should get a less noisy derivative.
I hope this helps.
Hitesh
Software Dev. Engineer
Software Dev. Engineer