On Thu, June 14, 2012 10:35 am, alan.b.pearce@stfc.ac.uk wrote: >> >I would not try to use the internal clock to do USB, even if it is >> >possible on the chip (and I don't know if it is). >> > >> >USB requires better clock accuracy than you are likely to get without a >> >resonator or crystal. >> >> That's what I've always thought as well. But don't recent FDTI USB >> chips work >> without an external crystal? >> > > There are a handful of Microchip USB chips that claim to have an internal > oscillator stable enough to work with USB, but I don't believe that chip > is one in this category. > > The only one I can think of that I have discovered claims to be stable > enough is the PIC24FJ64GBxxx series. To meet the USB spec (full speed, at least), you need a pretty accurate oscillator, I believe about 0.5%. In general, the standard internal oscillator just won't cut it across temperature. The internal oscillator is often accurate enough for USB operation at room temp, but you run the very real possibility of it drifting out of compliance with the USB specification. Low speed USB allows much looser tolerances on USB, and I believe it runs within the USB specification across the temperature range of the most parts (-40 to +85). There is a future part (announced and available "real soon") that does USB clock recovery (PIC16F14xx) - n= o external (timing) components needed. Matt Bennett Just outside of Austin, TX 30.51,-97.91 The views I express are my own, not that of my employer, a large multinational corporation that you are familiar with. --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .