PXIe-6345 Differential Voltage Measurement Noise Issue

[Fais94]

Hi everyone,

I’m using an NI PXIe-6345 to measure a single Li-ion battery cell voltage. I want 10 Hz data, so I’m sampling at 100 S/s and planning to decimate/average down to 10 Hz.

The problem is that I’m seeing very high noise (~20 mV) on the measurement, which is far above the device spec (expected noise is tens of μV).

For comparison, I’m doing the same measurement with an NI cDAQ, and the noise there is two orders of magnitude lower, which suggests this isn’t a wiring issue.

My current wiring setup:

• AI+ → battery positive

• AI– → battery negative

• AI GND → ??? (I’m unsure if AI– should be shorted to AI GND)

I’m using differential input mode.

Could anyone advise:

1. How exactly should I wire a single floating battery cell for differential measurement?

2. Should AI– ever be shorted to AI GND?

3. Any tips on reducing noise in this setup?

Thanks in advance for any guidance!

[ZYOng]

See Table 1 in Field Wiring and Noise Considerations for Analog Signals

Use Single Ended Referenced mode, or add bias resistors to use Differential mode.

[Matt5]

Thank you for the suggestion and sharing the resource!

We have now tried the measurements again by adding bias resistors (with R=10 kOhm, R=50 kOhm, or R=100 kOhm):

1. Resistors from AI+ to AI GND and from AI- to AI GND
2. Resistor only from AI- to AI GND

Unfortunately, the addition of bias resistor(s) did not improve the noise in the signal. We also tried the RSE method (Input signal wired to AI+ and AI GND), but the noise in the signal was even worse for that case.

Any other thoughts or suggestions would be greatly appreciated!

Could there possibly be something wrong with the Analog-to-Digital converter or other internal component of the PXIe-6345 module, causing more noise in the measurement?

[santo_13]

---

@Matt5 wrote:

Thank you for the suggestion and sharing the resource!

We have now tried the measurements again by adding bias resistors (with R=10 kOhm, R=50 kOhm, or R=100 kOhm):

1. Resistors from AI+ to AI GND and from AI- to AI GND
2. Resistor only from AI- to AI GND

Unfortunately, the addition of bias resistor(s) did not improve the noise in the signal. We also tried the RSE method (Input signal wired to AI+ and AI GND), but the noise in the signal was even worse for that case.

Any other thoughts or suggestions would be greatly appreciated!

Could there possibly be something wrong with the Analog-to-Digital converter or other internal component of the PXIe-6345 module, causing more noise in the measurement?

---

Please share exact connections and wiring photos if available.

If the Lithium cell is unloaded, you shouldn't be observing a high noise in the signal unless it is picked up from the environment through wiring.