On Tue, Jan 2, 2018 at 3:33 AM, David C Brown wrote: > I have tried the RC filters suggested and they do not improve the situati= on. > > For completeness I ought to say that the cable from the ICD2 to the targ= et > uses one of those little RJ to single in line adaptors. There is about > two inches of flat cable from the ICD2 to the adaptor then six inches of > individual wires to the target. The only opportunity for cross talk i= s > in that two inches of flat cable. The Resistor and the 2 capacitors should be as close as possible to the microcontroller, rather than the ICD2. There is six inches of wire again to= the target right, which makes it a total of 8 inches wire ? Try placing the res= istor and the capacitors no more than 1 or 2 inches away from the target. > > Wouldn't a corrupt WinUSB.dll cause problems with other USB devices? An= d Not necessarily, only userspace USB communication I/O would be broken. (This is a relatively slow communication method, in comparison to the kerne= l space driver. Normal USB applications are unlike to use userspace communications.) Note that the data from the kernel space driver is copied to the userspace driver. > can I simply replace WinUSB.dll to test this theory.? > Possibly. IIRC MPLAB brings it's own drivers right ? There could be multiple paths to different versions of the same library name under windows. So need to look for such oddities. It is better to uninstall the WinUSB driver from within MPLAB and reinstall it again through the same process. It might take a bit time to search for different copies of the library with different names there are different names to the same thing. I have one at my end named STLinkWinUSB Happy Snorkelling, Manu --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .