USB 6008 performing random resets

[dsavir]

Hi all, 

We are using the USB 6008 to control two motors and acquire data from a number of motion sensors. We are controlling it with LabVIEW 2010 on Windows 7 on a laptop. Usually it works fine, but sometimes for no reason we can see the USB-6008 resets, causing a TRUE output to all channels, causing the motors to whir round and round until stopped.

It once happened three times in one day, and then didn't happen for a week. It once happened when sensors were activated and sometimes not.

Does anyone have any idea how to prevent this, or alternatively how I might check for this in the software and stop it if it happens?

Thanks,

Danielle

[crossrulz]

USB?  Windows 7 introduced a "power saving feature" where it will randomly turn off USB ports.  If you go into Device Manager, you can turn off this feature for your ports.

From a safety perspective, you should not be using a USB DAQ for control.  Something like that belongs in a cRIO or FPGA card.

[SnowMule · Accepted solution]

6008's are fine for control, as long as proper safety measures are put in place. 

Yes, they can reset on occasion... or if windows feels like it... or if windows crashes, or applies an update and reboots the machine... any number of reasons.

And yes, all the digital outputs get pulled high when that reset happens.  Analog output goes to 1V, you can use a comparator and some logic gates to drive the digital outputs to a "safe" state for your application, then enable the digital outputs by driving the analog output to 5V.

A more elegant solution, as crossrulz mentioned, is using a more capable device like cRIO or PXI controller.  Obviously this adds considerable cost to a system.

Power savings options are usually the culprit.  Turn that off on the root hubs and each of the devices.

Are you driving the 6008's through a USB hub, or straight off the PC?  Is the PC rebooting or throwing errors (Check the system event logs)?

[dsavir]

Thanks, CrossRulz and SnowMule!

CrossRulz, the power saving feature was indeed activated, we turned it off. I thought the power management was to turn the USB off if it wasn't in use, and the resets happened when the USB was in use, so... Windows doesn't like hardware?

SnowMule. thanks for the safety circuit suggestion. If I turned off the power management, will Windows still reset the USB? As this system is simple and the lab has the USB600 changing to a CRIO or PXI is not practical at this time.

Thanks again!

Danielle

[johnsold]

Danielle,

Regardless of the cost of the system, if a glitch - whether produced by the OS, the user, or a power failure - can cause damage or danger, then it is imperative that a suitable safety mechanism be incorporated. The cost of one hospital visit due to a limb injured by a runaway motor will be far higher than the cost of incorporating safety in the design. 

The kinds of things SnowMule suggested may be sufficient and have low material costs.  If you do not know how to design the required safety features, consult your Terry Pratchett quote and get professional help.

Lynn

[dsavir]

Lynn, thank you for the warning. Thankfully, the system is not that dangerous, so I think that the solution snowMule suggested will be sufficient.

Thank all for your help!

Danielle