Channel setup sample mode

Forum for discussing PicoLog
Post Reply
Indi2Go
Newbie
Posts: 0
Joined: Thu Jan 07, 2021 9:29 am

Channel setup sample mode

Post by Indi2Go »

Hello,
in the help file I can find following statement:
"Sample mode - average of multiple readings over the specified sample interval; or latest sample only"

I have an 5000 series oscilloscope, what does this setting mean when I set the sample mode to "average" and an sample interval of 5ms (200Hz).
How many samples are then averaged? Can the count of averages samples influence this somehow? (the 'raw data rate'?)

How is the resolution (8/12/14/16 bit) set for this channel?

Thx,
Stefan

Ps.: sorry if this was already answered in this forum, but I was not able to search properly. I was not able to search for the term "channel setup". I always get result where one of this both words a occure but not the entire phrase.
(I used the + sign, nothing worked....)

Gerry
PICO STAFF
PICO STAFF
Posts: 1145
Joined: Mon Aug 11, 2014 11:14 am

Re: Channel setup sample mode

Post by Gerry »

Hi Stefan,

To answer your questions:

1/ The way PicoLog 6 acquires samples is to sample faster than the rate that you set (unless you're sampling at the maximum permitted rate). So, when you set the Sample Mode to "Average" and use a sample interval of 5ms what happens is that it will capture as many samples as possible, e.g. 5 at 1ms and then average out the values. If you set the Sample Mode to "Latest Value" it will give you the last of the 5 values. This means that you can specify a sample rate (average), but also get a feel of what your data will look like with a faster sample rate than the one you specified (latest value). Note that the maximum sample rate is 1kS/s, so at 1kS/s averaging and latest sample will produce the same result. A useful feature is that, if you select a different sample mode for 2 channels of a PicoScope or Data Logger, you effectively get 2 sample rates at the same time, e.g. if you want to look at some of the variance as well as the mean or want to look at a clean and dirty (noisy) signal together. If you select Latest Sample, as you won't be capturing the first 9 samples you can't know the true variance, but with noise you will have a good enough approximation.

So lets look at 2 fast changing (for a data logger) signal captures demonstrating how the mean and variance would look for that specific signal. In the data files below, you can see that there is a Pulse Waveform being generated by the AWG of a PicoScope, and using the Sample Mode of Latest Sample the True peak of the pulse is captured. This is because the sample interval is small enough in comparison to the pulse width (also the edges of the waveform are so fast as shown in the Sig Gen capture that the probability of a sample point not being either at the pulse peak or base line voltage are negligible). So, in this specific case the peaks of the pulse waveform are always present, but for fast signal waveforms in general this will not always be the case. When using the Sample Mode of Average for the capture, you can see that there is a reduction of the pulse peaks, which is effectively a rounding of the sharp edges of the pulse (seen in the Sample Mode Latest Value capture). So, this is effectively a crude filtering of the high frequencies in the signal. In PicoLog 6, you can only use Math Channels to average the same sample point across waveforms when more than one waveform has been captured, so Sample Mode Average is a useful way of averaging data, to find mean values or reduce high frequencies along a waveform.
AWG one fifth cycle.png
Sample Mode Latest Sample - Zoomed out.png
Sample Mode Latest Sample - Zoomed in.png
Sig Gen edge of Sq wave.png
Sample Mode Average - Zoomed out.png
Sample Mode Average - Zoomed in.png
So, let's look at a temperature data logging example, that combines the two sample modes on 2 channels of the same device. As we are on lock-down I'm working from home, and I don't have access to a TC-08 data logger, so I have used scopes to simulate the sort of slowly changing temperature variation you can get accompanied by a lot of noise generated in the typical data logging of a thermocouple. in the data files below you can see that there is a regular ramping down of the data (from the Averaged data in green). The start of the 2nd peak would be wrong if only the Latest Sample data was captured, while the extremes of high and low temperature would be wrong if only the Averaged values were captured.
Simultaneous Latest Sample and Average TC-08 simulation.png

Note that using a 'Sample Mode' of 'Average' in PicoLog 6 also increases the effective resolution of the signal, and is similar to the effect of the averaging process when using 'Resolution Enhancement' in PicoScope 6. For further information, you can find a comprehensive discussion on this here: topic40938.html?&p=145370&hilit=latest+sample#p145370).

2/ If we consider a 10ms sample interval, then whether it takes 10 at 1ms or 2 at 5ms is going to give essentially the same average, but clearly a different Latest value.

3/ You can only set the Resolution for the whole Scope as shown below:
Setting Scope Resolution.png
In the Channel options you can increase the effective resolution by selecting the amount of Resolution Enhancement you want as shown below:
Setting channel Resolution Enhancement.png
Regards,

Gerry
Gerry
Technical Specialist

Post Reply