How can I use LabVIEW to ground floating SCXI-1520 channels (in order not to lose scanning capabilities from the multiplexer)?

[mlindsey]

Currently I am reading quarter bridge strain with several SCXI-1520s. When in MAX I noticed my values were no longer reading after a certain gage and as I attempted to troubleshoot, found out that the multiplexer in this module will not scan past any open channel (or if it does, channel results are not consistent) for a very logical reason. If I lose a gage, I usually lose any data from any subsequent gages which we really cannot afford.

Are there any suggestions programming in LabVIEW (right now i'm using NI-DAQmx with an analog task set up)so that each time my loop executes, the program will check for any open channels and either not collect that data or change the order of acquisition (i.e. place the open channels last on the scan list?

I'm open to anything that might be of help 🙂

[Laura F.]

Hi mlindsey,

By open channel, do you mean that one of the strain gauges has broken and no longer is outputting a signal?

When you say that you lose data, what exactly do you mean? Do you get data, but the values are incorrect? In Measurement and Automation Explorer, how many channels were you scanning when you noticed this behavior?

Thanks,
Laura

[mlindsey]

Yes, by "open channel" i mean that there is no signal returning from the gage because it has broken off the structure or the signal to the terminal block has been disconnected somehow.

When I lose data, I'm referring to the data that is acquired on following channels. The behavior appears to be fairly random (which was a troubleshooting nightmare 🙂 ) but usually if a channel is open, then that and those behind it in the scan order do not return any data. It returns a number like 10.357E-6 and I do not pick up any noise or fluctuations. As far as I understand, the multiplexer stops scanning at this point (open channel) because it has nothing to ground to and subsequently "hangs" - not (or randomly) scanning the following channels. I have gone through the bridge configuration and as of yet, have not figured out how to fix this using any hard-wiring techniques. I emailed NI, and was sent a library file to "ground the channels" but it actually resets the Analog input Task I have defined each loop.

I am new to NI-DAQmx so I do not know if this will alter any configurations I already have in MAX for NI-DAQmx. Will it override? I am concerned it will slow my loop time, and also I cannot see the code for this sub.vi so am a little skiddish about using it if i do not know if it can be tailored to my app.

What do you think? Am I on the right path?

[Laura F.]

Hi mlindsey,

What seems to be happening once your strain gauge becomes open is that the signal then goes out of range and saturates your amplifier. There are two solutions to this problem. The first is once an out of range value is detected, stop the task and remove that channel from the task so that you no longer scan it. I think you have already stated that you are concerned about slowing down your data acquisition if you implement this technique. The second option is to use an M Series DAQ board as the MIO board that is cabled to your SCXI system. This issue has been addressed with the M Series boards.

Please let us know if you have additional questions.

Thanks,
Laura

[mlindsey]

Laura -

The system I am experiencing this with is using a PCI-6052E, but the system in planning right now uses a PCI-6259. From what you posted earlier - I understand that this issue will be resolved with the 6259 card?

ml

[johnsold]

If you are not using the full range of the DAQ input device you could put a resistive voltage divider parallel to the strain gage. Select the resistors so that if the sensor is disconnected you get a voltage that is out of the range of your valid data but still within the input range of the DAQ device. And select high values so that they introduce as little error as possible while the sratin gage is connected. If the resistors are mounted close to the DAQ device, they are unlikely to have the broken connection problems. This will introduce some error into the measurements, but since you know the values of the resistors and they do not change, you should be able to compensate for the errors. Clamping the inputs to a voltage within the input range via diodes might also work.

Lynn

[mlindsey]

Lynn -

From first glance, what range of error do you think would occur in the measurements? (using a 120 Ohm gage with a voltage divider circuit)

ml

[Laura F.]

mlindsey,

The 6259 is an M-series board and it will take care of this issue.

Thanks,

Laura

[devchander]

mlindsey,

can I plz get a feedback wheather using a 6259 Mseries card solved your problem?,

plus, can you tell me if SCXI 1520 was acquired in parallel or Multiplexed mode with 6259?

i had a similiar set of drawback with SCXI 1520 when used with PXI 6071 E series card.

but the solutions we used  included appropriate filter settings and  Hardware grounding of Open channels via a labview program. This did give good results with a few compromises on Maximum sampling rate we could use.

but i would like to know if switching from Eseries to M series DAQ card can solve all these problems?

thanks in advance.

[Ryan_V[DE]]

devchander,

I can't speak for mlindsey, but I would like to confim what Laura has suggested.

Regards,