Drift in PCI 6221 analog channels

[mmfaleiros]

    Hi,

    The VI depicted below performs the measurement of three analog channels. Then it calculates (Si - So)<Bi>/So*B,

where Si = Channel Ai1/Channel Ai2

and Bi = Channel Ai0

    When Si = So, the average measure should be zero, but in fact what I get is the behavior showed in the plot below.

     I use PCI 6221, BNC 2110 and LabVIEW 7.1.

     What can be causing this drift? Could you help me to correct this?

     Thank you,

     Marcelo.

[Abel_Souza]

Hi,

Your VI don´t have a create channel VI, so problably you are reading anything, because you don´t configure your acquisition. You can see examples of data acquisition on LabVIEW Find Examples.

Regards,

Abel

NI Brazil

[mmfaleiros]

I'm sorry, I should explain it better.

This is just a section of the code. Three channels are reading three batteries of different voltages. When S = So, however, the average must be zero.

My question is: this drift is caused by the pci6221/bnc2110 electronics?

The required create and config channels are posted below.

[John_P1]

I'd be more interseted in your signal connections--what input terminal configuration are you using?  If Differential or NRSE, are you using bias resistors (see Field Wiring and Noise Considerations)?

Also, the graph of (Si - So)<Bi>/So*B complicates things a bit, can you just plot the voltages of the batteries directly?

[mmfaleiros]

The batteries plot is the following.

They seem OK, but the problem is that the value of each measure changes at the fourth decimal place, always increasing. But what I need is the result of the calculation showed in the first post, which is of the order 10E-4, the same order of the fluctuation in each channel.

I'm using differential input configuration. The bias resistors, in the suggested range, minimize but don't cancel the effect.

Thanks.

[John_P1 · Accepted solution]

Are you seeing about a ~40 uV drift on the ±10V range of the card?

This would be well within the accuracy specs of the 6221.  However, it does sound like the drift occurs over time.  If you leave it connected long enough do the voltages continue to drift or do they stabilize?

It could very well be that the battery voltages themselves are what is drifting.  However, if adding bias resistors helps then it's possible you are accumulating small amounts of charge on the ADC--what resistor values did you try?  You might also consider switching to RSE and connecting the battery (-) terminals to the common AI Ground to avoid having to use bias resistors.

Best Regards,

[mmfaleiros]

Yes, the batteries was drifting. But it was not the only problem. I had a bad contact in the AI GND.

Fixed the bad contact and replacing the batteries by a voltage supply, the problem was solved.

The result without drifting is depicted below.

Thank you very much by the help.

Marcelo