HELP computer crash when using data translation with labview

[Sundown]

Hi,

So to explain what I'm doing for starters, I have used DT-Link with DataTrans devices and Labview in the past without any issue. Mainly D/O blocks, A/Os and maybe 1 A/I. So I'm relatively familiar with working with these.

I began working on a new project that involes a dt9817 using 18 or so D/Os and then they had purchased a 9812(10v) for the 4 A/Is needed. I didn't seem to have any problems setting up the D/Os, in a while loop I referenced the 9812 and put in a write function. Now when I created the A/Is I did it just like I had before for a single A/I. I spent a little time figuring out how to get them to read from the correct port, but eventually figured it out.

The problem is now that sporadically when the program is running, the computer will go to blue screen and then restart!! The only significant things I have time to see before it restarts are:

DT9810_x64.sys

bluescreen error

code: 50

I know it has something to do with DT Link as it is almost the only thing in the program and it only goes to bluescreen when the program is running.

To answer questions:

Brand new computer. Asus i7 quad core.

I have tried to repair LV Link multiple times with no success.

Finally, I'm assuming it is based around a problem with the way I'm sampling the 4 A/I channels, but I can't think of any other way to do it. If some one can help me figure this out, or post an example code for multiple A/I channels into xy graphs I would really appreciate it.

Thanks in advance,

Luke

[Kira_T]

Hi Sundown,

I have a couple more questions about your set up.  What operating system are you running?  Is it 32bit or 64bit?  What version of LabVIEW are you running and is it 32 bit or 64 bit?  What version of DT-Link are you using?  Is any of this different than previous setups you have used?  Thanks!

Regards,

Kira T

[Sundown]

Windows 7, 64-bit computer. Labview is the most current, we just bought it. I did just realize that Labview is running as 32bit, so I don't know if this can cause issues or how I can go about changing that. (As a side note, the computer has been running slow when using Labview, could this be the cause?)

Update: I did finally get a hold of some tech people at Data Translation. I told them the problem and what errors I was seeing. They were fairly confident that the problem was an unhandled error in their driver and not with Labview. So they are currently looking into the issue, and I'll post and let you know when I hear back from them!! Thanks!!

[Sundown]

I just wanted to update this thread. Since having this error I worked out a few things that seem to have made the error more infrequent, however, it is still causing a major issue when the chemists are trying to run and a blue screen is pulled.

I have gotten to an executable on a few projects and I have seen the error come up on both the development computer and target computers. It has been narrowed down to a problem with the digital outputs, and the LV Link digital output driver. Unfortunately the Data Translation people have been less than helpful, so I was hoping that someone might be able to post some example code to work from.

This is the crash dump file from the last time I got a blue screen. If it makes a difference, it was taken from a computer with an executable.

Here is the code that pulls the error.

 

Thank you to anyone who can help. I am really stumped at this point.

[Sundown]

sorry, the last part of the dump file is:

[RavensFan]

I'm not familiar with the device you are using.  But looking at your code, it appears you are creating a resource with the DatX subVI's that have the sun on it.  Then you are writing to the device.  I don't see any subVI where you are closing the resource.  So if you are doing this over and over again in a loop conitnually creating resources, you may be running out of memory.

In a normal LabVIEW application that uses DAQmx, you would create a task, use the task, and then close the task.  If you are doing this repeatedly, then the creation should occur before the loop, and the close should happen after the loop.  Only the repeated process of using the task would occur in the loop.

If that VI you have with a sun on it is a "creation" type of subVI, then I would not put it in a loop.

[Sundown]

That hadn't occured to me, but I will try adding in a close block and see what happens. I will post when I see something! As the error is very intermittant I don't know when this will be.

Thank you very much for your fast response,

Luke

--edit--

As a side note, I remember the Data Translation people telling me that the program should automatically close the call when a new one is created. So I don't know if this will fix the issue or not, any other suggestions would also be appreciated!

[Sundown]

Unfortunately after some testing I had the blue screen error show up twice. This was with the close task now in line with the do's If there is anything else anyone can suggest, I would be greatful.

Thank you,

       Luke

[RavensFan]

Were the create task and close task inside or outside the loop?  Don't open and close repeatedly in the loop.  Open once before the loop and close once after the loop.

[Sundown]

Yes they are. As far as I can tell they have to be opened inside the loop because I am trying to write to 3 different channels, so whenever I want to switch to writing to a different channel a new task has to be created, correct?

If anyone out there has an example of a multi-channel DO with Data Translation, that would be a big help!