One AI channel influences another

[bios2000]

Hello,

I'm using a NI PCI-6251 measuringcard to get the analogue output of two distance-sensors. The output is on the range of 0...10 V and I connected it to AI 0 and AI 1. As samplerate I use the triggersignal of my sensors (200us high, 15.2ms low).

Now, if the first sampled signal changes its value, the second sampled signal is influenced by it (see the first picture). If I add a third analogue channel to read between the reading of AI 0 and AI 1, the interference disappears (see the second picture).

How could this be explained?

Much thanks!
Carsten Frey

Message Edited by bios2000 on 01-09-2009 09:13 AM

[SFK]

---

bios2000 wrote:

Hello,

I'm using a NI PCI-6251 measuringcard to get the analogue output of two distance-sensors. The output is on the range of 0...10 V and I connected it to AI 0 and AI 1. As samplerate I use the triggersignal of my sensors (200us high, 15.2ms low).

Now, if the first sampled signal changes its value, the second sampled signal is influenced by it (see the first picture). If I add a third analogue channel to read between the reading of AI 0 and AI 1, the interference disappears (see the second picture).

How could this be explained?

Much thanks!
Carsten Frey

Message Edited by bios2000 on 01-09-2009 09:13 AM

---

Carsten, hello,

you are experiencing a phenomenon called "signal ghosting". It is caused by the ADC capacities not being able to completely reach the new voltage potential of the second multiplexed channel in the given sampling time.

A detailed explanation is given here: How Do I Eliminate Ghosting From My Measurements?

You might also be interested in this article, as it describes how to prolong the conversion time: How is the Convert (Channel) Clock Rate Determined in NI-DAQmx and Traditional NI-DAQ?

Best regards,
Sebastian