pico measurements produces large noise

[Yuval_Kuper]

Hi,

I'm writing a program that suppose takes measurements using the pico. the signal I'm measuring is signal that has steps in various voltage levels. Problem is, while measuring with the pico I get a really noisy measurement (about 0.8V), which doesn't appear if I take the exact same measurement using a regular oscilloscope.
I've tried using all the different average modes but the signal comes out really distorted.
What is the cause for this noise and how can I filter it without damaging the signal?

Thank you, Yuval

[Gerry]

Hi Yuval_Kuper,

You can go to this link for an explanation: https://www.picotech.com/support/topic40394.html?&p=143551&hilit=noise#p143551

Regards,

Gerry

[Yuval_Kuper]

Hi Gerry,
Thank you for the reply. I'm trying to find a way to do this by programming, and not through the oscilloscope's software. Is there any way to do so?

Thanks, Yuval.

[Gerry]

Hi Yuval_Kuper,

Sorry for the delay in answering (we have had a large work load followed by a change in working arrangements due to the Coronovirus).

You can implement what is being done in PicoScope 6 in program code. A filter is likely to impact your signal, so that would net be a good choice, which leaves you with:

1/ You can perform the Resolution Enhancement of PicoScope 6 if you're sample rate is sufficiently lower than the maximum sample rate for your PicoScope (so that you can over-sample the data) by applying a moving average filter to the data. What you need to do is decide on the width of the filter (i.e. the number of samples that it will use, better if it is an odd number so that the filter can be centred on a point) and then starting filtering at a distance of half of the (filter width - 1) from the start until half of the (filter width - 1) from the end. What you do, as a method, is step through each sample point and replace it with the average value of all of the samples within the filter width, then move to the next point and repeat.
The moving average filter will remove high frequency components, so in order to use it without changing the shape of the signal you need to over-sample the data (use a sample rate that is x * 5 * the signal bandwidth, where x is the (width of the filter - 1) see the text underneath the Res Enhance table at the bottom of the link here: https://www.picotech.com/library/oscilloscopes/resolution-enhancement.

2) You can perform waveform averaging if your captured waveform is repetitive. You just split the Overview Buffer that you declare memory for into multiple buffers. Then perform block captures of the repeated waveform into successive memory blocks, and finally average across the memory blocks, so that you are averaging sample 1 for blocks 1 to x, then sample 2 for blocks 1 to x, etc. where x is the number of block captures you have performed. Note that this will not change the shape of your waveform at all (it will just clean up the noise) but it will be limited by the number of samples per capture and the size of the Buffer Memory in your PicoScope.

Regards,

Gerry