Does using unity gain buffers on a differential floating signal convert the signal to a ground referenced differential signal?

[grimre2153]

Hello,

I am experiencing ghosting on my PXI-6225 card. I reviewed the manual and I am trying to implement the unity gain buffer suggestion. The issues I am running into is that I want to monitor multiple batteries using differential measurements. The battery is considered a floating source so normally I would use the connection recommendation for a floating signal source Differential measurement. However, I am having ghosting issues so I am implementing a pair of unity gain buffers. One for the positive terminal and one for the negative terminal of the battery. The buffers are connected to the power supply ground. Does this mean that the inputs to the DAQ are now Grounded signal source Differential measurements and should I used the connection recommendation for Differential Ground Referenced signal sources?

I am also having an issue with the the measurement on the DAQ floating above the actual battery voltage. I noticed that if I connect the AI GND to the GND of the unity gain buffer the signals look fine. Is this recommended? I'm assuming that the device and the DAQ all need to have the same ground reference.  I have attached an image of the circuit in question. Any support would be greatly appreciated.

Amir

[HollyBerry]

You will simply need to wire a pull down resistor between AI- to GND. Also, the power supply is just there to power the Op Amps and your measurement is still being powered by the battery. Therefore, you will want to continue wiring your system as a differential floating signal.

[johnsold]

The circuit you posted is meaningless. With the way you have drawn the grounds everything is single ended and the buffers are useless. 

I presume you really want multiple batteries, most of which will not have either terminal connected to ground?  Batteries are generally low impedance sources and the buffers should not be required. The buffers probably have higher output impedances than the batteries unless you have very small, very unusual batteries.

Please post a diagram of the battery circuit showing how the cells are connected and the points at which you wish to make measurements. Is any point in the battery stack grounded? What is the maximum voltage to ground in the battery? Do you need to protect your measurement system against transient events such as load dumps seen in automotive systems?

Lynn

[grimre2153]

Hello Lynn,

I am perform a constant load discharge using two batteries in series. The load is a 1K resistor. I want to monitor the battery as it discharges. The unity gain buffers came into play because I was seeing ghosting on other channels. The points of interest is the postive and negative terminals of the battery. The battery is not grounded. It is simply attached to a resistive load. The max voltage will around 4V. We may need to consider load dumps if we ever do anything more than constant load tests in the future.

Amir

[grimre2153]

This is the diagram.

[johnsold]

As Holly-S said in her earlier post, you need a resistor from AI- to GND and differential input connections.

What is connected to the other channels where you are seeing the ghosting? Those channels may be the ones needing buffers.

Lynn

[grimre2153]

Hello Lynn,

At the moment nothing is connected to the other channels. In the future the same constant load circuit will be connected to each of channel.

Amir

[johnsold]

You should ground, directly or through a resistor, any unsed inputs.

Lynn

[grimre2153]

Hello Lynn,

Thank you for you feedback.

Amir

[grimre2153]

Hi Lynn,

When performing the differential meausrement using the unity gain buffers, how do you maintain the same ground reference while measuring a floating source (a battery)? Should the ground of the unity gain buffers and the negative of the battery be connected using some sort of isolation?

Amir