OK, I think I've got it figured out. It looks like the USB framework
assumes you'll use a standard I/O pin for Vbus sense. But, the PIC32 has a
pin devoted to that function and a bit in a status register that indicates
whether Vbus is present. So, I changed this definition to this:


#define USB_BUS_SENSE U1OTGSTATbits.VBUSVD // Use this register bit to
detect bus voltage. hh 7/27/10

Then, at the top of my UsbTask(), I added this:

     if(OldVbus!=3DU1OTGSTATbits.VBUSVD){       // usb was plugged in or
unplugged
       OldVbus=3DU1OTGSTATbits.VBUSVD;          // remember the new value
       if(OldVbus){                          // newly attached
         USBDeviceAttach();                  // connect it
       }else{
         USBDeviceDetach();                  // disconnect it
       }
     }




Also, in UsbInit, I commented out the attach at the end. Instead, we'll
wait for the task itself to discover an attachment.

 // below commented out. We'll wait to attach in UsbTask.
 //  #if defined(USB_INTERRUPT)
 //      USBDeviceAttach();
 //  #endif
 }//end UsbInit




So, I had not dug into the details of the USB Framework. Instead, I
slightly revised the Generic USB sample code to do what I wanted. It
apparently left the USB attached even when no host was connected. It
appears that this resulted in a lot of bus errors when no host was
connected, really slowing stuff down due to all the interrupts. I'd
eventually get a watchdog timeout.

So... it seems to work so far. Time to test more!

THANKS!

Harold

--=20
FCC Rules Updated Daily at http://www.hallikainen.com - Advertising
opportunities available!
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.