Programatically recovering from a USB coms gitch

[cmc@uts.edu.au]

How can I programatically recover from a USB coms gitch using NI DAQmxBase 3.7 with a NI USB-6215 ? Unfortunately DAQmxBase Reset Device does not reset the device if there is a USB problem. I can remove the USB cable from the 6215, run DAQmxBase Reset Device and it returns no error !

Chris.

[JefeL]

Can you give us more information about your system? Are you running the applicaiton on Windows or Linux? What version of LabVIEW are you using?

[cmc@uts.edu.au]

I'm using LabView 2012 on Mac OS 10.9.2. I had the same problem with the previous version of DAQmx Base.

Chris.

[JefeL]

Would you be able to provide a simple test VI example that demonstrates the behavior? The fact that it doesn't return any error may or may not be expected behavior depending on the circumstances. If we can reproduce the issue we can find out more about why it is behaving in this manner.

If the root of the problem ends up being on the OS side of things we may still have options. The System Exec VI might allow you to perform command line manipulation of the USB driver to programatically unplug the device and then plug it back in.

[JÞB]

First off this is a good excuse to consider Windows and DAQmx full (NI Devmon can be configured to run an executable on device detection- You can write the execuatble to do anything including get the device in a known reset state.)

But that's beside the point.  USB "Glitches only happen if you fail to follow a few simple rules:

1 Always use an externally powered USB Hub of good quality.

2 Never let a User near your HUB

3 Use high quality USB Cables and those ferrites that came with the 6009.

[cmc@uts.edu.au]

Just run DAQmxBase Reset Device with the appropriate Device In ID and with the USB device plugged in with the green light flashing (as it should be). Initially DAQmxBase Reset Device returns no error as it should. Do this a few times to make sure - still no error. Then unplug the USB cable from the device and DAQmxBase Reset Device returns no error, when it should return an error.

Chris.

[cmc@uts.edu.au]

re: USB hub: 'm not going through any hub - this it straight from the device to a laptop.

re: cable: this is with the original NI cable that came with the 6215. We have tried different USB cables - unfortunately it does not fix the problem.

Chris.

[cmc@uts.edu.au]

Just noticed something curious. I repeated these tests using DAQmxBase Reset Device with a USB-6009 (using a different USB cable) and got the same results. I tried the lsdaq utility and it works correctly with both the 6009 and the 6215 - ie it correctly senses the presence of either device. However if I run the FWUpdate utility it recognises the 6009 but does not recognise the 6215. The 6215 is almost 5 years old but has never had a FW update (because most of the time it works fine). Why doesn't the FWUpdate utility recognise the 6215 ?

[JÞB]

---

cmc@uts.edu.au wrote:

re: USB hub: 'm not going through any hub - this it straight from the device to a laptop.

re: cable: this is with the original NI cable that came with the 6215. We have tried different USB cables - unfortunately it does not fix the problem.

Chris.

---

Oh, so you expect the cheapest port your laptop manufacturer could find powered by the cheapest power supply they could get to run occasionally to provide an "Industrial" solution?

I'm sure your iMac is nice for playing music.  Welcome to industry!

[cmc@uts.edu.au]

Hi Jeff,

Any progess with this issue ?

Chris.