# What could be the reason for these bands in this data acquisition measurement?

Here is zoomed view of 16 channel

The first plot looks very peculiar to me hence there are like two forbidden bands(looks like white lines as you see). This also appears if I plot any individual channel(again without any filter). For example this is for channel 7:

One can only see these white bands if the plot us plotted with dots.

What could be the reason for these white bands?

will be updated..

Your ADC should have a resolution of $$\\dfrac{20V}{2^{16}} = 305\mu\textrm{V}\$$

So if there are gaps in your data due to the resolution of the ADC we would expect them to occur every 300 microvolt or so.

Your filtered data looks like it has gaps about 1mV apart, which makes me suspicious of your filter software.

These white bands may be artifacts of the graphing software. I suggest zooming in on the voltage scale around one of them to see if you can estimate the size of the gap.

The small white bands are from the resolution of the ADC.

The large white bands are most likely from bits that are not there because of manufacturing defects in the ADC's that the DAQ card uses to measure the voltage. They look like this when looking at the ADC's bits vs voltage graph:

When the voltage 'lands' on that bit, it actually gets transfer to the bit above or below resulting in an error

• I have seen this behaviour on old BB hybrid's in the 70's when the state change of TTL current xxxxxx1111xx to xxxxx10000xx would shift the internal analog ground to Vref thus creating missing codes. Oddly it only occurred on Mil-std screen 883 parts and OK on Industrial ones supposedly lower quality. – Tony Stewart Sunnyskyguy EE75 Oct 8 '19 at 19:17
• They must be using an old/cheap ADC, maybe the multichannel ones have more errors? Most ADC's now tout 'no missing codes' – Voltage Spike Oct 8 '19 at 19:18