NI 5112 to 5114 DC Offset of 50-60 mV

[aparab]

Hi,

Till  date we used  NI 5112  PCI card  and have recently   tried to replace it with  a  NI 5114 PCi card.

The issue at hand is that we get a DC offset of 50-60mV with the  5114 Card.

The only change in the system is the card.

The  settings for the new card are same as the  5112.

Its hard to explain where the offset comes from &  how to eliminate it ?

any similar experiences!

[Nathan-P]

aparab,

Have you run Self-Cal?  Temperature depedent DC offset correction is a part of Self-Calibration.  

http://zone.ni.com/reference/en-XX/help/370592N-01/scopeviref/niscope_cal_self_calibrate/

Though since the two devices have different analog front ends, they will not behave exactly the same.  How exactly do you have the digitizer configured?

[aparab]

thanks,

I was on Vacation in between!

but  both the cards are calibrated.

the  software is the same.

we just i nterchange the cards & this offset Issue creeps us.

[Nathan-P]

What signal are you acquiring?  (amplitude, freqeuency, periodic/continuous, etc.)

Also, what settings are you using?  Which input range, input impedance, sample rate, etc.

What software are you using?  LabVIEW, Soft Front Panels, CVI?  

You confirmed that the cards are calibrated, but I'm assuming that is external calibration.  Self-Calibration and External Calibration are different.  External Calibration is done by a Calibration Lab, which needs to be done every 2 years for most digitizers.  Self-Calibration is a driver routine that the end user is expected to run to ensure the device is operating within the specifications by adjusting for temperature variations between its current setup and the temperature used in the Calibration Lab.

Please run Self-Calibration on the NI 5114, using niScope Self Calibrate.vi or in MAX click "Self-Calibration".

[aparab]

Thanks Nathan,

1. I am acquiring Amplitude Signals 0-1V

2. All the settings are  same for both the cards.

3. Labview

4.Self-Calibration  is done on 5114 card.

I will run the calibration again  &  come with more inputs

regards

Akshay

[Nathan-P]

---

@aparab wrote:

2. All the settings are  same for both the cards.

---

I understand this, but what are the settings you are using?  Accuracy for example depends on the vertical range selected.  

For all I know, you are measuring your 1V signal in the 40V range.  For example, the DC accuracy on the PCI-5114 of measuring 1V in the 10V range is:

±[(1.5% × |Reading - Vertical Offset| + 2% of Vertical Offset + 0.3% of FS + 280 µV)

or 1.5% * 1V + 0.3% * 10V + 280µV = 45.3 mV

This means that the expected voltage measurement will be 1V +- 45.3mV.  So the selected range for your measurement is important.

http://digital.ni.com/public.nsf/allkb/63E5D3110F9DE1BE86257497007BD063

[aparab]

thanks again Nathan ,

the thing is  other colleague is doing this I  am just overlooking the issue,  thts why i am short on  specifics.

I will check this out.

regards

akshay

[aparab]

hi,

well  the range is 1V & is same for both and other parameters  eg DC coupling , etc are the same(all these things r  in the  s/w )

we also do selfcalibration   but the offset value   50-60mv we get only for  5114

thanks ,

regards,

akshay

[Msengxxl]

Hello,

i am the engineer that found out that problem. Maybe i can give you some more facts:

The signal we are measuring is a sine signal with 700mV amplitude and a DC Offset of 2,5V.

The vertical range is set to 2V the vertical Offset to 2,5V. The Frequency is 20 Khz.

Coupling is set to DC. Input Impedance is set to 1MOhm.

And in fact there's a DC-Offset on both channels with 30mV-40mV.

We used to have passive HW-Lowpass filters in line, but bridging them has only less influence.

If we replace it with the old NI-5112 everything works fine...

[Nathan-P]

With your given setup, it appears the PCI-5114 is operating within its warranted DC accuracy:

DC Accuracy = ±[(1.5% × |Reading - Vertical Offset| + 2% of Vertical Offset + 0.3% of FS + 280 µV)

±[(1.5% × |700mV| + 2% of 2.5V + 0.3% of 2V + 280 µV)  =  ±66.7 mV

So since your measuring 700mV, what will actually be measured by the digitizer is 700mV ± 66.7mV.  If the measured result is less accurate than this, then the hardware will need to be returned for calibration/repair.  We only guarentee that the device will operate within this accuracy range.

Variation between digitizers is expected, escpecially between different models, since the analog path is different, as well as the ADCs.  Typically our digitizers will perform more accurately than our warrented specifications, since they are worst case limits.      

Regards,

Nathan