Yeah..... Try making that resistor Carbon Composition--you know the old style with carbon granules as the resistor media. They are noisy--which is why they are not used much these days--but this is a special case where you may need a little noise to kick the oscillator into motion. I know we have to use a 100K in series with our watch crystals, and still see start up delays of MANY seconds. K. At 11:35 AM 5/18/00, you wrote: >(Aside: the problem I had with the EEPROM access has been fixed. >Apparently, the "not really saving" bug bit me after I commented out >the #define EMULATED in the code. As such, the compiled code >was accessing the OSCCAL register instead of GPIO to talk to >the EEPROM (reg 0x05 is used for ICE-PIC). Silly, stupid mistake.) > >The current problem is now that the processor takes well over >5 seconds to start up from powerup. Power is coming up quickly, >and everything seems to be fine. > >I'm using a 12CE519 at approx. 150 kHz with 47 pF load caps. > >No series resistance takes over 5 seconds. > >Adding 10 kOhm of series resistance (OSC2-side) does not >permit the oscillator to start up. (FYI, resistor was 0805 SMT >1% film resistor.) > >I will try higher load caps (68 pF) tomorrow, once we get some >in. Before then, am I missing anything else? > >Thanks. >John > >-- >John Patrick >Software Engineer William K. Borsum, P.E. -- OEM Dataloggers and Instrumentation Systems & San Diego, California, USA