mega 2560 window 7 x64 crash

Has anyone run into issues with the 2650 crashing on shutdown with windows 7 x64, we have it on two different machines. Both when we connect and use a 2650 within a few mins of no activity the boards lock up. When we shutdown we get a BUGCODE_USB_DRIVER BSOD. We are using serproxy as well for use in flash.

Im not sure the best way to capture or post a dumpfile or what its even called to google for it. If there is something I can post to help solve the issue please let me know.

A Blue Screen of Death is always the result of either a bug in a kernel driver (or the kernel) or failing hardware. It is simply not possible for your Mega 2560 or SerProxy to be the culprit.

The first step is to identify the driver that caused the kernel panic. That tiny bit of information is extremely helpful. For example, if the video driver caused the kernel panic, updating the driver may resolve the problem.

Some things that may help...

• I know it is difficult to believe but Microsoft actually has some very good documentation describing how to troubleshoot a kernel panic. However, finding that documentation may be difficult. In other words, searching Microsoft's website may prove fruitless but it is still worth doing.

• An Event Log entry is added when a kernel panic occurs. Which log? What severity? I have no idea. I just know that you may find useful information somewhere in that mess.

• I believe an entry is added to the System Troubleshooter. Open Control Panel. Click System and Maintenance. Click Problem Reports and Solutions.

• If you find the right forum, there are people on the internet who will help you collect the relevant information and even help you identify the problem.

Well the two issues do go hand in hand and are in some way related to the device. I found prior posts about this same issue but it was on boards using FTDI drivers. But in my case its just an inf that points to the windows usb driver. We have two boards and the issue occurs in both on both of my dev machines (more or less the same) and a laptop all running win 7 x64. important to note the laptop is a clean windows install as well. Both mega's work fine for a short peroid of time and if I unplug and replug back in they are work again for a short time. But if I shut down with one attached I BSOD, if I shutdown without using it everything is fine.

Would it help to post a dumpfile?

Would it help to post a dumpfile?

I won't be able to do anything with it.

Does the problem occur if you have nothing connected to the board?

Do you have a HUB between the computer and the board?

What do you have connected to the board?

The only thing we use the board for is buttons that send a ground to digital inputs 2-4, but it will shutdown ok if we dont connect, once you connect (via sockets over serproxy) it will crash.

Yea the board sits on a hub (going to try switching that out, just assumed since i could connect the hub was fine).

ryanmills:
The only thing we use the board for is buttons that send a ground to digital inputs 2-4

Just 3 buttons? Nothing else?

Yea the board sits on a hub

Passive or powered?

Yep just three buttons, hub is passive (in the sense that its powered by usb).

It's also worth noting that the time out for crash seems random, i cant seem to isolate it to any one command/action, seems more to do with when its idle. Also environment is rather mild for temp and humility. I have two separate test beds both have the same issue.

ryanmills:
Yep just three buttons

Wired to ground? With the internal pull-up enabled?

in the sense that its powered by usb

That makes it a passive hub.

hub is passive

Test without it.

Yes, wired to ground, i'm not doing anything to set the pullup other than enabling digital reporting, setting it as an input then setting it to high. In my case im using as3glue. Would it help to post my as3 controller class? Did I miss a step? I did slap diodes on the ground to check for some sort of backfeed to ground, made to change.

Yes on the hub, just wanted to verify i was understanding correctly.

Going to be testing that first.