analog output spike

[bronson bunker]

Hello,

I am using the pci-6713 analog output board with a scb-68 block. The analog output channels 2 through 8 exhibit a short voltage spike at the start of the task. The task consists of running a small DC motor. This phenomenon does not occur in channel 1.

Cheers,

BB

[Jennifer O]

Hello Bronson,

Could you provide more detail about the spike that you are seeing?  Amplitude? Duration?

Does the spike occur any time you change the voltage, or is it simply in the power-up state (i.e. when they system is turned on, before a voltage is written to the channel in a test panel or an application)? The power-up voltage is specified at 0+/-200mV.

To avoid applying the power-up voltage to your motor, you can connect a relay between the analog output and your motor and use a digital line to switch the relay on after driving 0V to your analog line.

Hope this helps,

Jennifer O.
Applications Engineer
National Instruments.

[bronson bunker]

Alrighty,

I have connected the analog outputs to a PCI-6064E analog input. Output channels 0 and 7 operate normaly; channels 1, 2 and 3 exhibit a 1V ouput spike; and channels 4, 5 and 6 exhibit a 10V output spike (all regardless of the task voltage).The voltages of channels 2 through 6 drop to 0V after 1ms.

Connecting the analog outputs to an oscilloscope generates the same initial voltage spikes except that the output spike have the shape of a step of around 50ms in length (for channels 2 through 6). I do not know why there is a difference between the two measurement methods however both methods capture the same results when the output task is a simple square wave pulse at 0.2Hz (following the initial spike).

Cheers,

BB

[bronson bunker]

I forgot to mention that these spikes occur at the beginning of runtime..

[Jennifer O]

Hi Bronson,

I have set up a PCI-6713 to compare the output characteristics to what you are seeing with your board.  I was unable to see the voltage spikes you describe. Could you perhaps post a detailed description of how you are making your connections between the PCI-6713 and the PCI-6040E? Are you running an example program or an application that you have written? Could you post the code that you use or the name of the example? What is the amplitude of the square pulse you are writing?  Do you perform a write prior to starting your task?

I would like to test your exact setup to make sure that your board is working correctly and determine the cause of the spike.

Regards,

Jennifer O.

[bronson bunker]

The connections between the two SCB-68 blocks are: A0i to ACH0 and AOGND to AIGND (from the 6713 to the 6024E respectfully). I am using Matlab and Xpc Target to communicate with the I/O boards. I am not using any examples, instead I am using a Simulink models consisting simply of a square wave generator block connected to a rate transmission block connected to a the output block 6713. The generated wave has a 0.5 second period and an output of 2V however the wave output voltage has no effect on the output spike voltage mentioned above (a 1V wave generates the same results). I am not too sure what you mean by a write prior to my task but from what I can understand I am not writing anything prior to runing the task.

Hope this helps,

Eric

[Jennifer O]

At the time of the spike do you have any other hardware connected?  You mentioned that you are trying to control a DC motor.  I want to isolate the hardware (i.e. make sure your motor is disconnected) to make sure that the spike is not due to reflection.

When working with the DAQmx driver (in LabVIEW for example) it is common practice to write the data to a buffer and then start a task.  The data will not appear at the terminals until the start occurs.  If a start is not used the DAQmx write can be set to autostart. I'm not sure what calls the Simulink software makes to the driver.  I do not have this software to test my board as you do in your setup. Could you also remove the software from the equation by running two test panels in Measurement and Automation Explorer (MAX).  First start a test panel for analog input on your PCI-6040E and monitor it while you run a test panel for analog output on your 6713 (opening a second instance of MAX will enable you to use a second test panel).  If this reproduces the voltage spikes, would it be possible to post a screenshot? 

To this point I have assumed that you are working with the DAQmx driver.  It is also possible that you are working with Traditional DAQ driver.  You can verify your DAQ version by looking under the software tab in MAX.  Can you post the driver version you are using(i.e 7.x where x is a number)?  Also, are you using the National Instruments Simulation Interface Toolkit?  If so, what version?

Regards,

Jennifer O.

[bronson bunker]

Hello,

The spike occurs regardless of the hardware connected. At this point I have removed the DC motor and simply connected both PCI cards as mentioned in my previous message.

I have run the two test panels as you have suggested and found that all the outputs of the PCI-6713 were operating normally (no spikes). The PCI-6317 is working with the traditional DAQ driver version 7.0.1f4 and the 6024E is working with the DAQmx driver version 7.0.1f0. I am not using the simulation interface toolkit.

As both boards seem to operate normally under the test conditions using the NI software, there may be a problem with the Matlab software.

Thanks again,

Eric

[Jennifer O]

Hi Eric,

The spike may have been addressed in one of the more recent drivers.  Please try updating the driver to you are using with the PCI-6713 to either DAQmx or version 7.2 or later of the Traditional DAQ driver to see if it fixes the problem when communicating with your software.

Hope this helps,

Jennifer O.

[bronson bunker]

Hi,

   I have updated the drivers for the 6713 to the DAQmx as you have suggested and the outputs still exhibit the spike at the begining of the task. I have again run two test panels using the new software and see no problems in all the outputs.

  I have also tested the analog outputs of the the 6024E using the same Simulink model and setup as with the 6713 and found that there there were no abnormalities in the signal. If there is anything else I can do please let me know.

Many thanks,

Eric